PREDICTION OF HEADCUT MIGRATION USING A DETERMINISTIC APPROACH

Abstract

The rate of headcut migration is of specific interest for engineers designing earthen spillways and embankments.It is also of importance in landscape management and river geomorphology and restoration. The objective of this study wasto compare the measured rates of headcut migration from flume tests to predicted migration results from a simplisticdeterministic model. Headcut migration tests were conducted in a 1.8m wide and 29m long flume with 2.4m high sidewalls.Migration rates were predicted based on a simplified physically based deterministic equation using soil strength, soil unitweight, erodibility and critical shear stress, overfall height, backwater level, discharge, and hydraulic stresses in the plungepool region of the headcut. A total of 46 headcut migration tests are analyzed, 37 tests of a compacted CL material, 2 testsof a compacted SM material, and 7 tests of a compacted CL material overlying an erodible sand layer. The migration ratesvaried over 3 orders of magnitude from 0.02 m/h to 20 m/h. Factors such as soil strength and erodibility have a primary impacton the rate of headcut migration. Factors such as backwater and bulk unit weight have less of an impact. It was observedthat predictions of headcut migration in the high backwater tests were better correlated (r 2 = 0.92) than low backwater cases(r 2 = 0.42). This was attributed to the contrast in the behavior of the impinging jet in low and high backwater cases and presentunderstanding of stresses in this environment. The inclusion of physical parameters to account for soil strength, erodibility,hydraulic stresses, and force balance makes it possible to make physically based predictions of migration rate over severalorders of magnitude.