Annualmeeting:2017 CSDMS meeting-018: Difference between revisions
(Created page with "{{CSDMS meeting personal information template-2014 |CSDMS meeting first name=Francis |CSDMS meeting last name=Rengers |CSDMS meeting institute=USGS |CSDMS meeting city=Golden...") |
No edit summary |
||
| Line 7: | Line 7: | ||
|CSDMS meeting state=Colorado | |CSDMS meeting state=Colorado | ||
|CSDMS meeting email address=frengers@usgs.gov | |CSDMS meeting email address=frengers@usgs.gov | ||
|CSDMS meeting phone= | |CSDMS meeting phone=303-273-8637 | ||
}} | }} | ||
{{CSDMS meeting scholar and pre-meeting | {{CSDMS meeting scholar and pre-meeting | ||
| Line 14: | Line 14: | ||
}} | }} | ||
{{CSDMS meeting select clinics1 | {{CSDMS meeting select clinics1 | ||
|CSDMS_meeting_select_clinics1= | |CSDMS_meeting_select_clinics1=2) ANUGA - river flood morphodynamics | ||
}} | }} | ||
{{CSDMS meeting select clinics2 | {{CSDMS meeting select clinics2 | ||
| Line 28: | Line 28: | ||
|CSDMS meeting abstract submit=Yes | |CSDMS meeting abstract submit=Yes | ||
}} | }} | ||
{{CSDMS meeting abstract title temp}} | {{CSDMS meeting abstract title temp | ||
{{CSDMS meeting abstract template}} | |CSDMS meeting abstract title=Estimating model parameters necessary for simulating post-wildfire debris-flow timing | ||
}} | |||
{{CSDMS meeting authors template | |||
|CSDMS meeting coauthor first name abstract=Luke | |||
|CSDMS meeting coauthor last name abstract=McGuire | |||
|CSDMS meeting coauthor institute / Organization=University of Arizona | |||
|CSDMS meeting coauthor town-city=Tucson | |||
|CSDMS meeting coauthor country=United States | |||
|State=Colorado | |||
|CSDMS meeting coauthor email address=lmcguire@email.arizona.edu | |||
}} | |||
{{CSDMS meeting authors template | |||
|CSDMS meeting coauthor first name abstract=Jason | |||
|CSDMS meeting coauthor last name abstract=Kean | |||
|CSDMS meeting coauthor institute / Organization=USGS | |||
|CSDMS meeting coauthor town-city=Golden | |||
|CSDMS meeting coauthor country=United States | |||
|State=Colorado | |||
|CSDMS meeting coauthor email address=jwkean@usgs.gov | |||
}} | |||
{{CSDMS meeting authors template | |||
|CSDMS meeting coauthor first name abstract=Dennis | |||
|CSDMS meeting coauthor last name abstract=Staley | |||
|CSDMS meeting coauthor institute / Organization=USGS | |||
|CSDMS meeting coauthor town-city=Golden | |||
|CSDMS meeting coauthor country=United States | |||
|State=Colorado | |||
|CSDMS meeting coauthor email address=dstaley@usgs.gov | |||
}} | |||
{{CSDMS meeting abstract template | |||
|CSDMS meeting abstract=Debris flows pose a hazard to infrastructure and human life. However, predicting debris flows remains a challenge due to uncertainty in initiation mechanisms, and the difficultly in appropriately parameterizing the resistance equations that describe flow velocities. Additionally, one of the limitations to progress in modeling debris-flow timing is the lack of empirical data from natural watersheds that can be used for parameter estimation and validation of predictions. Most quantitative measurements of debris flows are conducted in flumes, or unique watersheds where debris flows are known to occur annually, both of which suggest particularly remarkable conditions that may not reflect the majority of conditions where debris flows are manifested. This research addresses those challenges by using measured debris-flow timing in nine watersheds that were burned by a wildfire in 2009 to calibrate and test debris flow model parameterizations. Debris-flow timing was captured using pressure transducers attached to the channel bed. We used a kinematic wave rainfall-runoff model that we developed in python using the landlab environment to model flow timing. We separated the nine study watersheds into two categories: calibration and testing. For the calibration watersheds, model parameters were estimated based on prior research and then changed iteratively using a storm with known rainfall to minimize an objective function of the observed and modeled flow timing. Following hundreds of model realizations, we arrived at a set of best-fit parameters for saturated hydraulic conductivity (Ks) and the Manning’s roughness parameter (n). We found that a single value of Ks could be used in each of the model watersheds because, following wildfires, this parameter is typically reduced to very low values with a relatively small variance. In contrast n varied systematically as a function of upstream contributing drainage area, and thus values of n could be estimated for uncalibrated basins. When Ks and n were applied to test basins without any calibration we found that a reasonable result in estimated debris-flow timing was attained. These results suggest that given the appropriate scaling estimates it may be possible to estimate debris-flow timing within minutes and to capture multiple debris-flow surges separated by several hours. | |||
}} | |||
{{blank line template}} | {{blank line template}} | ||
Latest revision as of 11:13, 27 March 2017
Browse abstracts
Estimating model parameters necessary for simulating post-wildfire debris-flow timing
[[Image:|300px|right|link=File:]]Debris flows pose a hazard to infrastructure and human life. However, predicting debris flows remains a challenge due to uncertainty in initiation mechanisms, and the difficultly in appropriately parameterizing the resistance equations that describe flow velocities. Additionally, one of the limitations to progress in modeling debris-flow timing is the lack of empirical data from natural watersheds that can be used for parameter estimation and validation of predictions. Most quantitative measurements of debris flows are conducted in flumes, or unique watersheds where debris flows are known to occur annually, both of which suggest particularly remarkable conditions that may not reflect the majority of conditions where debris flows are manifested. This research addresses those challenges by using measured debris-flow timing in nine watersheds that were burned by a wildfire in 2009 to calibrate and test debris flow model parameterizations. Debris-flow timing was captured using pressure transducers attached to the channel bed. We used a kinematic wave rainfall-runoff model that we developed in python using the landlab environment to model flow timing. We separated the nine study watersheds into two categories: calibration and testing. For the calibration watersheds, model parameters were estimated based on prior research and then changed iteratively using a storm with known rainfall to minimize an objective function of the observed and modeled flow timing. Following hundreds of model realizations, we arrived at a set of best-fit parameters for saturated hydraulic conductivity (Ks) and the Manning’s roughness parameter (n). We found that a single value of Ks could be used in each of the model watersheds because, following wildfires, this parameter is typically reduced to very low values with a relatively small variance. In contrast n varied systematically as a function of upstream contributing drainage area, and thus values of n could be estimated for uncalibrated basins. When Ks and n were applied to test basins without any calibration we found that a reasonable result in estimated debris-flow timing was attained. These results suggest that given the appropriate scaling estimates it may be possible to estimate debris-flow timing within minutes and to capture multiple debris-flow surges separated by several hours.
