Time Development of Scour in Clayey Soils by Submerged Vertical Circular Turbulent Impinging Jets

Abstract

Defining the erodibility of cohesive soils is an important task for designing hydraulic structures and many other applications where water and soil interact. The erodibility of these soils is typically measured, and of the available tests, the Jet Erosion Test, developed by Hanson and Cook (2004), is a widely-used in-situ technique. The test employs a submerged circular impinging jet to create scour in the soil, and the development of the scour hole with time is measured. The data from the first few hours of a test is then used to predict the scour depth at an infinite time, called asymptotic or equilibrium state. It is thought that if the scour hole is at equilibrium, the shear stresses on the soil surface will be equal to the soil’s critical shear stress. The critical shear stress is the shear stress on the soil's surface below which the soil would not erode. However, there are several issues with the analysis procedures for the Jet Erosion Test. First, the means of predicting scour depth at equilibrium by Blaisdell et al. (1981) appears to be overestimating equilibrium depths. Second, the estimates of shear stress on the bottom of the scour hole do not consider the impact of the scour hole shape on the jet’s velocity. Experimental measurements in both manufactured pottery clays and natural cohesive soils were to test various mathematical models for their ability to predict the time development of the scour hole and the equilibrium depth from scour depth measurements early in a test. Only simpler models of scour development were considered because such a model would be easier to adopt in practice by jet test users with a range of technical abilities. The models tested were a logarithmic model, a hyperbolic model by Briaud et al. (1999), and an exponential model by Sumer et al. (1993). The results were compared to the predictions of Blaisdell et al. (1981). The study used data from four experiments carried out for the current research and Cossette (2016), Amin (2016), and Mazurek (2001). For the present study, scour tests were conducted using vertical impingement conditions on each of two samples of two manufactured pottery clays (M390 and P300 pottery clays of Plainsman Clays Ltd. of Medicine Hat Alberta). The scour depth along the jet centreline, the maximum scour depth and its location, the radius of scour hole, and the scour hole's volume were measured after scouring times of 5 min, 10 min, 20 min, 30 min, 45 min, 1 h, 2 h, 4 h, 8 h, and every 24 h thereafter until equilibrium was reached. Once the scour hole had reached equilibrium, a point gauge and a laser displacement device were used to obtain the full scoured surface profile. It was seen that the model by Briaud et al. (1999) appeared to most frequently provide the best fit to the time development data from the scour experiments, although it did not perform the best in all cases. The use of the Briaud et al. (1999) model showed substantially improved estimation of the equilibrium scour depth over the Blaisdell et al. (1981) method. Using measured and estimated equilibrium scour depths, the critical shear stresses of several soil samples were estimated and compared to reported values. This was done using the original jet test approach laid out in Hanson and Cook (2004), a means of evaluating shear stresses on the soil suggested by Ghaneeizad et al. (2015), and another suggested means of assessing shear stresses by Mazurek and Gheisi (2009). For the original approach of Hanson and Cook (2004), it was found that the overestimated equilibrium depths by Blaisdell et al. (1981), which should produce underestimated critical shear stress values, are balanced by overprediction of the shear stresses on the bottom of the scour hole. The combination of these errors results in reasonable critical shear stress values. However, the combination of improved prediction of the equilibrium scour depths using Briaud et al. (1999), and estimation of scour hole bottom shear stresses using Mazurek and Gheisi (2009), produced the closest predictions of the critical shear stress to the reported critical shear stress values.