Methods for Analyzing Submerged Jet Erosion Test Data to Model Scour of Cohesive Soils

Abstract

HighlightsFifty-two jet erosion tests performed on four cohesive soils were analyzed by nine different methods.Nonlinear methods performed well on some individual tests but fit inconsistently overall.Several alternate linear solution methods outperformed the widely used Blaisdell method.Simple linear regression of erosion rate versus applied shear stress provided the most consistent relationship between erosion rate and critical shear stress parameters.Abstract. The submerged jet erosion test (JET) is widely used in lab and field settings to quantify erodibility of cohesive soils and determine erosion rate coefficients and critical shear stress values. Test devices with different scales and configurations have been developed in recent years, along with several alternative methods for processing the collected data to determine parameters of linear and nonlinear soil erosion equations. To facilitate standardization, 52 JET experiments were conducted on four different cohesive soils compacted at optimum water content and 2% dry and wet of optimum. Each test was analyzed using nine different methods, four based on the linear excess stress equation (including the commonly used Blaisdell method) and five based on nonlinear erosion equations, including two using the recently popular Wilson model. Results were analyzed to determine the erosion equations and parameter-fitting methods that most effectively represent the observed erosion rates and are of greatest utility for soil erosion modeling and the ranking and classification of soils according to erodibility. Methods based on nonlinear erosion equations fit some data sets well, but they exhibited poor correlation between the erosion rate coefficient and the threshold shear stress parameter for initiating erosion, which is problematic for soil erodibility classification work. Linear methods that simultaneously optimized erosion equation parameters to best fit the total depth of scour or the elapsed time needed to reach specific depths of scour performed better than the Blaisdell method, which has been the informally accepted standard of practice since the late 1990s. However, they also exhibited weak correlation of the erosion rate and critical shear stress parameters. Simple linear regression of average scour rate versus average applied stress provided an effective method for representing the erosion rate versus applied stress curve and exhibited the strongest correlation of the erosion rate coefficient and critical shear stress parameters. Keywords: Cohesive soil, Critical shear stress, Erodibility, Erosion, Erosion laws, Erosion models, Jet erosion test, Shear strss, Soil moisture.