Debris flow dynamics: A flume study of velocity and superelevation

Abstract

Debris flows are powerful, geophysical flows with the potential to cause fatalities and damage infrastructure. The hazard posed by these events is often related to the velocity of the flow. However, due to their spontaneous nature and rapid onset, opportunities to observe debris flows are rare. The aim of this investigation was to improve the understanding of debris flow dynamics in curved channels using a hardware laboratory flume model.
The purpose-built, variable slope, 8 m long flume enables a variety of debris flow boundary conditions to be studied. Here, 60 experimental tests were carried out at four slope of – 15°, 17°, 23° and 25°; for both straight (12 tests) and curved channels (48 tests). The modular construction of the flume enabled purpose-made channel curves to be inserted in the straight channel. Four channel bends with differing geometries were tested; three radii of curvature – 0.4, 0.55 and 0.7 m (for a bend angle of 40°) and one radius of curvature of 0.7 m for a 20° bend angle. For all runs the grain-size of the debris flow mixture consisted of a poorly sorted gravel/sand/clay mix (bulk density 1880 kg m-3, water content 17%). A combination of direct and video-based observations were used to record debris flow behaviour including variation in superelevation and the average and local velocity of the flow.
Channel gradient is shown to be an important factor in controlling debris flow velocity. The straight channel results demonstrate a strong linear relationship between debris flow velocity and channel gradient. Modelling debris flow hydraulics reveals a power-law relationship, suggesting Newtonian turbulent flow conditions. The magnitude of superelevation was strongly linked to debris flow velocity, radius of curvature and bend angle. Superelevation was shown to increase in a power law relationship with debris flow velocity, which was greatest for tighter bend geometries. A decrease in radius of curvature resulted in a non-linear increase in superelevation, while decreasing the bend angle promoted a decrease in superelevation. When compared to observed superelevation, predicted values from the superelevation equation can substantially overestimate by up to 210% in some circumstances, suggesting that predicted values and back-calculated velocity estimates should be regarded as maxima. Furthermore, calculation of ‘k’ from experimental data suggests that ‘k’ was not constant (varying by two-fold) and has a value less than ‘1’, contrary to the proposed theory.
This novel investigation has studied the influence of debris flow velocity, radius of curvature and bend angle upon superelevation, broadening the very limited knowledge base surrounding debris flow superelevation and its controls. For future studies it is recommended to: increase the number of experimental runs for each channel configuration; use a wider range of radii of curvature and bend angles; and improve the accuracy of the superelevation equation by introducing a parameter to account for bend angle, and to better define the significance and value of ‘k’.