PubAg

Main content area

Linking statistical and hydrodynamic modeling for compound flood hazard assessment in tidal channels and estuaries

Author:
Moftakhari, Hamed, Schubert, Jochen E., AghaKouchak, Amir, Matthew, Richard A., Sanders, Brett F.
Source:
Advances in water resources 2019 v.128 pp. 28-38
ISSN:
0309-1708
Subject:
estuaries, hazard characterization, hydrologic models, probability, statistical analysis
Abstract:
A method to link bivariate statistical analysis and hydrodynamic modeling for flood hazard estimation in tidal channels and estuaries is presented and discussed for the general case where flood hazards are linked to upstream riverine discharge Q and downstream ocean level, H. Using a bivariate approach, there are many possible combinations of Q and H that jointly reflect a specific return period, T, raising questions about the best choice as boundary forcing in a hydrodynamic model. We show, first of all, how possible Q and H values depend on whether the definition of T corresponds to the probability of exceedance of “H OR Q” or “H AND Q”. We also show that flood hazards defined by “OR” return periods are more conservative than “AND” return periods. Finally, we introduce a new composite water surface profile to represent the spatially distributed hazard for return period T. The composite profile synthesizes hydrodynamic model results from the “AND” hazard scenario and two scenarios based on traditional univariate analysis, a “Marginal Q” scenario and a “Marginal H” scenario.
Agid:
6374469