Variability in Soil Erodibility Parameters of Tigris Riverbanks Using Linear and Non-Linear Models

Abstract

Most researches have predicted soil erosion of cohesive riverbanks using linear (excess shear stress model) and non-linear (Wilson model) models based on two soil parameters (detachment coefficient, kd, and critical shear stress, ?c) of the linear model and two soil mechanistic parameters (mechanistic detachment parameter, b0, and threshold parameter, b1) of the non-linear model. The goal of this research was to quantify the soil erodibility parameters of Tigris Riverbanks on Nu’maniyah-Kut Barrage reach using linear and non-linear models through the model parameters at three different water contents: dry side, optimum side, and wet side of water contents. Soil samples were collected from three locations south of Baghdad city on Nu’maniyah-Kut Barrage reach of Tigris Riverbanks. Six soil samples acquired from these sites were laboratory tests achieved using a miniature version of Jet Erosion Test device (“mini” JET) to determine the erodibility parameters of both linear and non-linear models. Blaisdell solution (BL) and scour depth solution (SD) were applied to determine (kd and ?c) of linear model from JETs data. Physical soil characteristics; including bulk density, particle size distribution (sand%, silt%, and clay%), average particle size (D50), and angle of repose were reported for six samples acquired from the three sites. The results showed lower value of kd of toe in compared with bank side for some specific sites as observed for both BL and SD solutions of excess shear stress model especially at wet side of water content. No general pattern of ?c related to different water content were observed. The parameters (b0 and b1) of non-linear model have the same behavior of linear model parameters (kd and ?c), but with different magnitude related to different water contents, respectively.