granular effects in debris flows

Granular effects in debris flows are usually assessed by dimensionless numbers, such as numbers of Savage, Bagnold, and Iverson, which measure the relative significance of granular interaction, and the values indicate that the granular effects are generally ignorable. But observations suggest robust phenomena pertain to grain composition in many ways. This implies that the dimension analysis does not apply to the recognition of granular behaviors in debris flows, partly because we have not really a direct description of changes in grain compositions of debris flows. We have proposed and confirmed that for debris flows the material grain size distribution (GSD) satisfies a unified function, P(D) = C*power(D, – μ)*exp(–D/D_c), where P(D) is the exceedance percentage of grains beyond size D (mm), and C, μ, and D_c are parameters, with a semi-log relationship between C and μ. Then the grain composition is characterized by the GSD parameters μ, and D_c, respectively representing the fine and coarse content of the materials. In this study we present a variety of appearances to illustrate how grain compositions impact on the initiation, formation, motion, and deposition of debris flow. Results indicate that debris flow occurs through a selection mechanism in which soil or sediment blocks of different grain compositions initiate in different ways and form separate surges in different flow regimes. The flow properties (X), such as the speed, the discharge, the density, are all dependent on the GSD parameters in power laws: X ~ power(μ, –m) and X ~ power (D_c, n); and the power laws impose constraints on the fluctuation of the dynamical quantities. In particular, the GSD evolves from the randomly aggregated grains to the fluid with some self-organized constitute.