Flow of variably fluidized granular masses across three-dimensional terrain 2. Numerical predictions and experimental tests

Citation
Rp. Denlinger et Rm. Iverson, Flow of variably fluidized granular masses across three-dimensional terrain 2. Numerical predictions and experimental tests, J GEO R-SOL, 106(B1), 2001, pp. 553-566
Citations number
22
Categorie Soggetti
Earth Sciences
Journal title
JOURNAL OF GEOPHYSICAL RESEARCH-SOLID EARTH
ISSN journal
21699313 → ACNP
Volume
106
Issue
B1
Year of publication
2001
Pages
553 - 566
Database
ISI
SICI code
0148-0227(20010110)106:B1<553:FOVFGM>2.0.ZU;2-I
Abstract
Numerical solutions of the equations describing flow of variably fluidized Coulomb mixtures predict key features of dry granular avalanches and water- saturated debris flows measured in physical experiments. These features inc lude time-dependent speeds, depths, and widths of flows as well as the geom etry of resulting deposits. Three-dimensional (3-D) boundary surfaces stron gly influence flow dynamics because transverse shearing and cross-stream mo mentum transport occur where topography obstructs or redirects motion. Cons equent energy dissipation can cause local deceleration and deposition, even on steep slopes. Velocities of surge fronts and other discontinuities that develop as flows cross 3-D terrain are predicted accurately by using a Rie mann solution algorithm. The algorithm employs a gravity wave speed that ac counts for different intensities of lateral stress transfer in regions of e xtending and compressing flow and in regions with different degrees of flui dization. Field observations and experiments indicate that flows in which f luid plays a significant role typically have high-friction margins with wea ker interiors partly fluidized by pore pressure. Interaction of the strong perimeter and weak interior produces relatively steep-sided, flat-topped de posits. To simulate these effects, we compute pore pressure distributions u sing an advection-diffusion model with enhanced diffusivity near flow margi ns. Although challenges remain in evaluating pore pressure distributions in diverse geophysical flows, Riemann solutions of the depth-averaged 3-D Cou lomb mixture equations provide a powerful tool for interpreting and predict ing flow behavior. They provide a means of modeling debris flows, rock aval anches, pyroclastic flows, and related phenomena without invoking and calib rating rheological parameters that have questionable physical significance.