Prediction of the Backwater Level Due to Bridge Constriction in Waterways

Abstract

Worldwide, bridges and culverts built across rivers are obstacles to the flow which cause an increase in water depth at the upstream of the structure that significantly intensifies flooding of land and property upstream. Therefore, it is important to understand the effects of bridges and culverts on water levels for flood damage reduction, flood risk management, scour evaluation, flood risk mapping, and maintenance of rivers and channels. Moreover, the methods available in the literature are generally not convenient for engineers due to the complexity of the equations and procedures of each method. On the other hand, the simple methods available in the literature do not contain a high level of accuracy. Hence, acquiring a simple accurate empirical method for computing backwater is necessary. In this study, a series of parametric studies is conducted to examine the influence of different factors on backwater. The results of the parametric studies along with multiple regression analysis are used in deriving a simple accurate mathematical model for computing backwater. The proposed method is firstly compared with the most commonly used method, energy method, for different skew angles and roughness cases. The comparison of the results of the proposed and energy methods indicates high correlations between the two methods. Furthermore, the proposed method is validated by comparing its results with experimental data for normal (at 0º), and skewed crossings at 30º and 45º. The overall absolute average percentage difference between the proposed method and experimental data is found to be 5.1%, while the overall root-mean-square error is found to be 0.008. Thus, the empirical method proposed by this study is considered highly accurate as well as simple in comparison with available methods in the literature. Additionally, the proposed method is applicable for rectangular and arch bridges, multiple opening bridges, and any type of crossings (normal and skewed) in compound channels.