Back to Search Start Over

Enhancement of semi-theoretical models for predicting peak discharges in breached embankment dams

Authors :
Chao Wu
Jianmin Zhang
Bo Wang
Wenjun Liu
Zhenyu Wu
Xin Liu
Yong Peng
Yunliang Chen
Sha Yang
Source :
Environmental Fluid Mechanics. 20:885-904
Publication Year :
2020
Publisher :
Springer Science and Business Media LLC, 2020.

Abstract

During the application of semi-theoretical models to predict peak discharges for breached embankment dams, it is often encountered to be lack of necessary model parameters such as breach height. In order to solve this deficiency and improve the application of the semi-theoretical models for estimating the peak discharge, a mathematical relation is proposed to predict the breach height (Hb) using the water depth above breach bottom (Hw) based on a subset of a composite database (72 historical dam failures). The breach height relation has a high coefficient of determination (R2) when it is applied in the composite database. Using the prediction by the breach height relation as a substitute for the observation, these semi-theoretical models used here produce similar results in terms of both prediction accuracy and uncertainty. Moreover, the dispersion of the flood peak discharge predicted by forecasting models is weakened and the skewness is improved by adopting the breach height relation. Therefore, the application of semi-theoretical models to predict peak discharges can be improved obviously. So, it is found that the breach height using the proposed relation can be a satisfactory substitute when the observed value of Hb is unavailable in a dam failure. Besides, the prediction accuracy will decrease slightly with the increase of database, but the decrease is limited. It should be noted that the proposed breach height relation is only suitable for predicting the height of embankment dam breach in principle.

Details

ISSN :
15731510 and 15677419
Volume :
20
Database :
OpenAIRE
Journal :
Environmental Fluid Mechanics
Accession number :
edsair.doi...........7cf4238fd97ceeefcdb119be345bdcd5
Full Text :
https://doi.org/10.1007/s10652-019-09730-9