Numerical Modeling of Flood Wave Behavior with Meandering Effects (Euphrates River, Haditha-Hit)

Abstract

A Numerical model for routing flood in a meandering rivers is presented. It is based on modified form of the complete one-dimensional Saint-Venant equations of unsteady flow. These equations are modified such that flows in the meandering river channel, left overbank floodplain, and right overbank floodplain are all identified separately. Thus, the difference in hydraulic and geometric properties and flow-path distances are taken into account for all these three divisions of the valley cross-section. The numerical model was applied to the Euphrates River in its reach between Haditha Dam and Hit city along 124.4 km to make a sensitivity analysis of the effect of meandering on the following parameters: the river peak discharge, peak water level, lag time of peak discharge, and lag time of peak water level along the river reach under study for various values of Manning roughness coefficient of the floodplain. Six cases were taken to compare the effect of meandering in the above parameters, cases A1, A2, and A3 are made for straight river and cases B1, B2, and B3 for meandering river. For both cases, three values of Manning roughness coefficient were given for the flood plain, 0.05, 0.07, and 0.1 respectively. The corps of engineers computer programs HEC-RAS 3.1.3 was used to determine the above parameters from a given flood caused by a hypothetical foundation failure of Haditha Dam. A comparative analysis indicates that the peak flow of the flood wave in Hit city for case B1 is 11.2 % more than for case A1 under the effect of meandering when the flood wave moves between Haditha Dam and Hit city. It is found that the meandering of the river had increased the peak flow of the flood wave at Hit city for case B2 by 13.6% over the peak flow of case A2, and had increased the peak flow of case B3 by 15.1% over the peak flow of case A3. The time lag between the start of failure of Haditha dam and arrival of the peak flow to Hit city is 14:45 hours for case A1 and 12:10 hours for case B1, this means that the meandering of the river had reduced the time of arrival of the peak flow to Hit city by 2:35 hours comparing cases A1 and B1. The meandering of the river was reduced the time of arrival of the peak flow to Hit city with 3:10 hours and 3:30 hours XIV when comparing cases A2 and B2 , A3 and B3 respectively. Comparing cases A1 and B1 showing that the maximum flood wave elevation in Hit city for case B1 is higher than that in case A1 with 0.97 m, and its time of arrival is lesser by 2:20 hours. Comparing cases A2 with B2, and A3 with B3 showing that the maximum flood wave elevation in Hit city for cases B2, and B3 are higher than that in cases A2, and A3 with 1.02 m, and 1.08 m respectively, while its time of arrival is lesser with (2:55), and (3:50) hours respectively. Dimensionless parameters: roughness ratio (nr), discharge reduction (Dr), and stage reduction (Sr) with other parameters such as time lag of peak discharge (TLD) and time lag of peak level (TLL) were developed to express graphically the relationship between the different cases of the flood wave routing of the case study. The dimensionless parameters: discharge reduction (Dr) and stage reduction (Sr), and the parameters of time lag of peak discharge (TLD) and time lag of peak level (TLL) were all increased with the increase of the dimensionless parameter roughness ratio (nr). The one-dimensional numerical model that used in this study was verified using a two-dimensional model over a selected small meandering reach in the study area. The two-dimensional model software (RMA-2) was used to calculate the velocity in two directions, water surface elevation, and discharge along the small meandering part in this reach. The results show a good agreement between the two-dimensional and the modified one dimensional models