Investigation of Hydraulic Jump Characteristics in Rough Beds Using M5 Model Tree

Hydraulic jumps are frequently used for excessive kinetic energy dissipation under hydraulic structures. The main objective of this study is to investigate the potential of M5 Model Tree for predicting the hydraulic jump characteristics, namely roller length and sequent depth based on Carollo et al.’s (2007) experimental data. A comparison of statistical results between M5 Model Tree, artificial neural network (ANN) and empirical-based equations of the sequent depth ratio (h2/h1) shows that M5 Model Tree allows a simple and reliable estimate for the sequent depth ratio. Also, the roller length could simply be calculated in rough bed without determining the bed characteristics and type of the jump. The results show that the classification of hydraulic jump in rough beds is different from smooth beds. It is beneficial investigating the hydraulic jump with different rough bed and Froude number to develop a more complete understanding the classification of hydraulic jump in rough beds.