Evaluation of cohesive soil erosion parameters
MetadataShow full item record
Assessment of cohesive soil erosion is of paramount importance in various areas of science and engineering. For example, soil erosion can cause increased pollution in water bodies and scour around bridge piers, which is the primary cause of bridge failure in the United States. Erosion is often modeled using an excess shear stress approach, where excess shear stress is the shear above the critical value for initiation of erosion (τc). This approach requires evaluation of τc and also an erodibility coefficient (kd). Although there is a range of exponents used for this relationship, in many studies erosion is calculated as a linear function of excess shear. Generally the erosion parameters τc and kd, as well as the exponent for excess shear, if the linear assumption is not used, must be determined for local soil conditions on a site-specific basis. The most widely used in situ technique to measure soil erodibility is the Jet Erosion Test (JET). However, almost all the available empirical relationships for erosion assessment have resulted from flume experiments where the flow is surface-parallel. This approach is not appropriate for the JET because the shear stress calculation based on mean flow characteristics diverges from the turbulent measures in the JET. This important aspect has been overlooked in developing the JET methodology, resulting in underlying assumptions that are physically inconsistent with the actual conditions of the test. In this study, data from both JET tests and flume experiments using the same soil sample were analyzed to determine the extent to which the two approaches produce consistent estimates for the soil erosion parameters. A numerical model also was used to extend the JET results and provide a more complete representation of shear stress. Results show that the estimates for erodibility parameters using the JET apparatus can vary widely and are associated with a wide range of uncertainty. In addition, τc estimates using the JET are about 20 times less than values obtained from the flume tests. Moreover, results show that the linear erosion law is incorrect and it can lead to large errors in the estimated erodibility parameters. These findings suggest the need for a reassessment of the JET method used for characterizing cohesive soil erosion. Additional modifications are required to correctly characterize stresses applied by the jet to the soil surface. This research highlights the inadequacy of widely accepted methods for cohesive soil erosion assessment and calls for a re-evaluation of underlying assumptions of the JET method.