2018 CSDMS meeting-090: Difference between revisions
Tanghui0822 (talk | contribs) (Created page with "{{CSDMS meeting personal information template-2018 |CSDMS meeting first name=Hui |CSDMS meeting last name=Tang |CSDMS meeting institute=Un |CSDMS meeting city=Tucson |CSDMS me...") |
Tanghui0822 (talk | contribs) No edit summary |
||
| (5 intermediate revisions by 2 users not shown) | |||
| Line 2: | Line 2: | ||
|CSDMS meeting first name=Hui | |CSDMS meeting first name=Hui | ||
|CSDMS meeting last name=Tang | |CSDMS meeting last name=Tang | ||
|CSDMS meeting institute= | |CSDMS meeting institute=University of Arizona | ||
|CSDMS meeting city=Tucson | |CSDMS meeting city=Tucson | ||
|CSDMS meeting country=United States | |CSDMS meeting country=United States | ||
|CSDMS meeting state=Arizona | |CSDMS meeting state=Arizona | ||
|CSDMS meeting email address= | |CSDMS meeting email address=huitang@email.arizona.edu | ||
|CSDMS meeting phone=5403156320 | |CSDMS meeting phone=5403156320 | ||
}} | }} | ||
| Line 18: | Line 17: | ||
}} | }} | ||
{{CSDMS meeting select clinics2 2018 | {{CSDMS meeting select clinics2 2018 | ||
|CSDMS_meeting_select_clinics2_2018= | |CSDMS_meeting_select_clinics2_2018=3) Data for natural hazards | ||
}} | }} | ||
{{CSDMS meeting select clinics3 2018 | {{CSDMS meeting select clinics3 2018 | ||
| Line 29: | Line 28: | ||
|CSDMS meeting abstract submit=Yes | |CSDMS meeting abstract submit=Yes | ||
}} | }} | ||
{{CSDMS meeting abstract title temp2018}} | {{CSDMS meeting abstract title temp2018 | ||
{{CSDMS meeting abstract template 2018}} | |CSDMS meeting abstract title=Physics-informed machine learning approach for predicting post-fire debris flows based on thresholds | ||
}} | |||
{{CSDMS meeting abstract template 2018 | |||
|CSDMS meeting abstract=Post-fire debris flow is a common hazard in the western United States. However, after decades of efforts in the debris flow research community, universally applicable post-fire debris flow predict methods are still lacking. Large discrepancies in the post-fire debris flow initiation mechanism are the main source that limits the predictive accuracy of debris flow. Improve and understanding these discrepancies is significant to possibly improve the debris flow modeling. In this work, we propose a data-driven, physics-informed machine learning approach for reconstructing and predicting debris flows. By using a classic supervising modern learning technique based on logistics regression, the logistics regression functions are trained by existing direct field measurements and debris flow numerical simulations from Las Lomas after 2016 Fish fire and then used to predict debris flow in different drainage basin where data are not available. The proposed method is evaluated by two classes of simulations: sediment transport model and runoff model. In runoff simulations, five drainage basins are considered: Las Lomas, Arroyo Seco, Dunsmore 1, Dunsmore 2, Big Tujunga. In sediment transport model, Las Lomas and Arroyo Seco watersheds are applied. Excellent predictive performances were observed in both scenarios, demonstrating the capabilities of the proposed method. | |||
}} | |||
{{blank line template}} | {{blank line template}} | ||
Latest revision as of 22:01, 18 May 2018
Log in (or create account for non-CSDMS members)
Forgot username? Search or email:CSDMSweb@colorado.edu
Browse abstracts
Physics-informed machine learning approach for predicting post-fire debris flows based on thresholds
Post-fire debris flow is a common hazard in the western United States. However, after decades of efforts in the debris flow research community, universally applicable post-fire debris flow predict methods are still lacking. Large discrepancies in the post-fire debris flow initiation mechanism are the main source that limits the predictive accuracy of debris flow. Improve and understanding these discrepancies is significant to possibly improve the debris flow modeling. In this work, we propose a data-driven, physics-informed machine learning approach for reconstructing and predicting debris flows. By using a classic supervising modern learning technique based on logistics regression, the logistics regression functions are trained by existing direct field measurements and debris flow numerical simulations from Las Lomas after 2016 Fish fire and then used to predict debris flow in different drainage basin where data are not available. The proposed method is evaluated by two classes of simulations: sediment transport model and runoff model. In runoff simulations, five drainage basins are considered: Las Lomas, Arroyo Seco, Dunsmore 1, Dunsmore 2, Big Tujunga. In sediment transport model, Las Lomas and Arroyo Seco watersheds are applied. Excellent predictive performances were observed in both scenarios, demonstrating the capabilities of the proposed method.
