On the relevance of numerical simulations to booming sand

We have performed a simulation study of 3D cohesionless granular flows down an inclined chute. We find that the oscillations observed in [L.E. Silbert, Phys. Rev. Lett., 94, 098002 (2005)] near the angle of repose are harmonic vibrations of the lowest normal mode. Their frequencies depend on the contact stiffness as well as on the depth of the flow. Could these oscillations account for the phenomena of "booming sand"? We estimate an effective contact stiffness from the Hertz law, but this leads to frequencies several times higher than observed. However, the Hertz law also predicts interpenetrations of a few nanometers, indicating that the oscillations frequencies are governed by the surface stiffness, which can be much lower than the bulk one. This is in agreement with previous studies ascribing the ability to sing to the presence of a soft coating on the grain surface.