 |
| Titel |
SENSITIVITY ANALYSIS AND BAYESIAN INFERENCE OF MANNING'S N FRICTION COEFFICIENT DURING A TSUNAMI EVENT USING POLYNOMIAL CHAOS |
| VerfasserIn |
Ihab Sraj, Kyle Mandli, Clint Dawson, Ibrahim Hoteit |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250087117
|
| Publikation (Nr.) |
 EGU/EGU2014-1134.pdf |
|
|
|
|
|
| Zusammenfassung |
| We present an efficient method to perform a sensitivity analysis of the Manning’s N friction
coefficient in the event of tsunamis. We also infer Manning’s N friction coefficient using
water surface elevation data obtained during Tohoku tsunami. We characterize the Manning’s
friction coefficient by three different parameters assumed constant in three regions: N1
on-shore, N2 near-shore and N3 deep-water. The efficiency of our approach stems from the
use of polynomial chaos expansions to build an inexpensive surrogate for the numerical
tsunami GeoClaw model that can be used to perform the sensitivity analysis. The surrogate
also reduces the computational burden of the Markov Chain Monte-Carlo sampling
needed for the Bayesian inference. Our objective is to sharpen three initial estimates
of the three uncertain parameters. Our results indicate that Manning’s N friction
coefficients have a Maximum-A-Posteriori (MAP) values of N2=0.011 and N3=0.185
while for N1 no meaningful MAP value can be determined using the available data. |
|
|
|
|
|