Debris flow impose major hazard in mountainous cities like Hong Kong which can cause catastrophic events. Debris flow protection in form of rigid barriers with deflector mounted on the top is recommended in the current design guideline in Hong Kong as prescriptive measure only, and there is no compelling experiments or literature to justify its functionality.

Scaled experiments have been carried out using dry uniform sand to interact with rigid barrier deflectors in a 5 m long flume channel. The channel is with a rectangular cross section and inclined at 26º from horizontal. The deflector angle measured from horizontal varies from 0º (orthogonal), 30 º, 45 º, and 60º. The effective height measured vertically from the tip of deflector to the channel bed varies from 66 mm, 105 mm,...[ Read more ]

Debris flow impose major hazard in mountainous cities like Hong Kong which can cause catastrophic events. Debris flow protection in form of rigid barriers with deflector mounted on the top is recommended in the current design guideline in Hong Kong as prescriptive measure only, and there is no compelling experiments or literature to justify its functionality.

Scaled experiments have been carried out using dry uniform sand to interact with rigid barrier deflectors in a 5 m long flume channel. The channel is with a rectangular cross section and inclined at 26º from horizontal. The deflector angle measured from horizontal varies from 0º (orthogonal), 30 º, 45 º, and 60º. The effective height measured vertically from the tip of deflector to the channel bed varies from 66 mm, 105 mm, 125 mm, and 144 mm respectively as the deflector angle increases. The adopted relative grain size is 0.012, and target Froude number is at around 4. A calibrated DEM model has also been used for parametric studies from varying length of deflector, effective height, and the channel inclination to observe their influence on the interaction mechanism.

Both experimental and numerical results reveal a ramp-like dead zone form while the flow mass impacts the barrier. The steepness of the dead zone is directly proportional to the effective height of the barrier, so as to the impedance effectiveness on the peak overflow velocity and maximum launch length. For deflector angles equate to or be greater than 45°, the approaching flow mass would be impeded with evident dissipating mechanisms. For deflector angles are less than 45°, a shallow ramp-like dead zone would be formed which increases both the launch velocity and launch length. For deflector angles greater than 30°, the launch angles of overflow are upward towards downstream which indicates an evident dissipating mechanism. A 0° deflector would increase both launch velocity and launch length by approximately 40% comparing with the case of without deflector. A 60° deflector would reduce launch length and launch velocity by approximate 20% and 30% respectively comparing with the case of without deflector.

The numerical parametric study reveals that the reduction factor concept from recurve wall in coastal engineering may not be applicable to measure the impedance effectiveness of the rigid barrier with deflector in the case of dry granular flow. This is deal to the flow mechanisms of sea wave and dry granular flow are representing two distinctive flow nature of either purely continuum or discrete. It is also revealed that increase in the length of orthogonal deflector would incur shielding effect which increases both the launch velocity and launch length. This phenomenon could supplement to the recommendation from Kwan 2012 which suggested a lower bound length of the deflector only.