Author(s): Urvashi Malani; Sahita I. Waikhom; Sanjay M. Yadav

Linked Author(s):

Keywords: Total load transport; Sand bed rivers; Multi-linear regression; Buckingham’s π theorem

Abstract: Sediment transport is a complex phenomenon. Being a dynamic process, it varies from year to year. Behavior of sediment transport is studied; many researchers and models are developed for prediction of sediment load using probabilistic, shear stress, stream power and energy slope approach. Present study aims to develop a total load transport model for sand bed rivers. It is crucial to consider particle size as it governs the phenomenon of sediment transport. In order to develop model, multi-linear regression approach is used to consider the effect of different influencing parameters of the phenomenon. Buckingham’s π theorem is used to find out influencing parameters with total load (QT) as dependent parameters and depth of flow (D), width (B), slope (S), mean flow velocity (v), fall velocity (ω), shear velocity (U*), Density (ρ), gravitational acceleration (g), mean particle diameter (d50) and specific weight of sediment (γs) are used. Dimensionless π terms are formed using U*, D and ρ as repeating variables. Ratio of some terms are interchanged by considering the physical phenomenon of sediment transport after obtaining independent dimensionless terms. To find the influencing parameter and develop model, data from 2 gauging stations Hoshangabad and Barman are chosen from Narmada River, India. From regression analysis ranking was given to each parameter at both sites and then vS/U*, ρU*²/Dγs, vs/√gd50 and D/d50 are chosen as influencing parameters. Multi-linear regression model is developed considering the logarithmic values of influencing parameters considering 70% of the data. Remaining 30% data is used for calibration and testing of model. Model was tested on 5 G.S. of Narmada River and results showed average score of 70% with discrepancy ratio ranging from 0.5 to 2. Well-known total load predictors for sand bed rivers like Yang [16, 18] were used to predict total load.

DOI: https://doi.org/10.1007/978-981-97-6009-1_21

Year: 2022