Diffusion–compaction coupling controls pore-pressure dynamics in granular–fluid flows

Eric C.P. Breard, Claudia Elijas Parra, Mattia de’ Michieli Vitturi · 2026

Read Diffusion–compaction coupling controls pore-pressure dynamics in granular–fluid flows by Eric C.P. Breard, Claudia Elijas Parra, Mattia de’ Michieli Vitturi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-18462.txt.

Introduction

Breard, Elijas Parra, and de' Michieli Vitturi show diffusivity in granular-fluid mixtures is not intrinsic but emerges from pore-pressure diffusion plus granular compaction. From two-phase mass balance for a soft gas pile, they derive excess pore pressure changes over time. At the thin-flow, small-pressure limit, this reduces to a coupled diffusion equation with a source set by pore share changes. Modal analysis splits drainage from compaction, and a unit-free source-to-diffusion ratio unifies values across nearly two orders of magnitude of bed height.

Catalog record

CREATOREric C.P. Breard, Claudia Elijas Parra, Mattia de’ Michieli Vitturi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE123605
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.18462

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Diffusion–compaction coupling controls pore-pressure dynamics in granular–fluid flows Eric C.P. Breard, Claudia Elijas Parra, Mattia de’ Michieli Vitturi Abstract Excess pore pressure in granular–fluid mixtures can transiently suppress frictional contacts and dramatically enhance flow mobility, yet its evolution is commonly modeled using constant effective diffusivities. Here we show that the apparent diffusivity is not intrinsic but emerges from the coupling between pore-pressure diffusion and granular compaction. Starting from two-phase mass conservation for a deformable, gas-saturated granular assembly, we derive an evolution equation for excess pore pressure that captures deformation of the granular skeleton. In the thin-flow, small-excess-pressure limit, this reduces to a one-dimensional diffusion–compaction equation with a time-dependent source term controlled by porosity changes. A modal analysis yields a reduced basal equation that separates diffusive drainage from compaction-driven forcing and identifies the corresponding timescales. This framework introduces a dimensionless source-to-diffusion ratio, Ψ 0 \Psi_{0} , which governs the competition between these processes and collapses effective diffusivities obtained from high-resolution two-fluid simulations over nearly two orders of magnitude in bed height. This scaling implies that the apparent diffusivity, and thus flow mobility, is not intrinsic but depends on flow thickness through the competition between diffusion and compaction. Incorporating this physics into a depth-averaged model demonstrates that the resulting closure reproduces the thickness dependence of pore-pressure decay and runout observed in experiments. These results provide a physically grounded description of pore-pressure evolution in granular–fluid flows and clarify how diffusion–compaction coupling controls their mobility. I Introduction In geophysical settings, gravity-driven granular flows such as volcanic pyroclastic density currents (PDCs) and debris flows can attain unexpectedly long runout distances because of excess pore-fluid pressure. These mixtures of solid particles and interstitial fluid (gas or water) may develop high pore-fluid pressures that temporarily reduce intergranular friction, thereby partially fluidizing the material and enhancing its mobility ( 21 ) . In debris flows, for example, sudden compaction of a water-saturated sediment mass can generate pore pressures exceeding hydrostatic equilibrium when permeability is low, liquefying the flow’s interior ( 21 ) . Likewise, concentrated pyroclastic density currents have been observed to propagate as a basal grain-rich avalanche cushioned by pressurized gas, distinct from the dilute turbulent ash cloud above. The presence of excess pressurized pore fluid in such concentrated granular currents is now recognized as a key mechanism for reduced frictional resistance and extreme mobility ( 27 ) . A fundamental driver of this behaviour is the coupling between granular bulk volume change (dilatation or compaction) and pore-fluid pressure diffusion. As a granular flow deforms, particles rearrange and either create void space (dilatation) or collapse voids (compaction). Compaction tends to squeeze the pore fluid (liquid or gas), elevating local pore pressure if the fluid cannot escape quickly, whereas dilatation produces suction. This dilatancy–pore-pressure feedback, central to soil mechanics and debris-flow dynamics ( 21 ) , has since been demonstrated in controlled experiments and numerical models. Two-phase continuum simulations have shown that internal pore pressure is dynamically modulated by intermittent episodes of dilatation and compaction within a moving granular mass (known as pore pressure feedback, 22 ), explaining transient basal pressure signals observed in small-to-large-scale experiments of granular media ( 8 ; 28 ; 20 ) . The competition between pore-pressure generation (from compaction) and dissipation (via Darcy-type diffusion) gives rise to characteristic timescales. If diffusion is slow relative to deformation, excess pressures are sustained and the material remains highly mobile; if diffusion is rapid, the flow quickly re-solidifies as pore pressure equilibrates. 18 formalized this interplay by introducing two key timescales—one for downslope advection and one for pore-pressure relaxation—and showed that their ratio, along with the material’s initial dilative tendency, governs whether a surge remains fluidized or quickly transitions to a frictional regime. This framework generalizes earlier shallow-flow models by 17 , which treated pore-pressure evolution as an advection–diffusion process within a depth-averaged granular mixture. Several material properties and initial conditions control the effectiveness of this diffusion–compaction coupling. Grain-size distribution strongly influence the pore-pressure diffusion rate: fine-grained, polydiperse size particle distributions generate low-permeability beds that dissipate pressure slowly, sustaining fluidization over longer timescales ( 29 ; 33 ; 34 ; 9 ) . Laboratory studies of initially fluidized granular columns show that thicker beds retain elevated pore pressures far longer than thinner ones, consistent with a diffusive timescale t_{d}\sim H^{2}/D . However, the inferred diffusivities still vary systematically with initial height, suggesting that D is not a constant material property in these systems ( 29 ; 35 ) . Laboratory studies of initially fluidized granular columns show that thicker beds retain elevated pore pressures far longer than thinner ones, consistent with a diffusive timescale t_{d}\sim H^{2}/D . However, the inferred diffusivities still vary systematically with initial height, suggesting that D is not a constant material property in these systems ( 29 ) . Continuum modeling of dam-break configurations with incompressible solvers—where the solid volume fraction is assumed constant—exhibits a similar inconsistency: reproducing the observed runout distances requires prescribing an effective diffusivity that depends on the initial bed dimensions. In such models, high-aspect-ratio columns tend to lose pore pressure more rapidly during collapse, whereas shorter columns remain pressurized over much of their evolution, producing nonlinear dependencies between runout and initial column height ( 1 ) . Taken together, these observations indicate that a classical Darcy description with constant diffusivity is often inadequate. In static defluidization experiments, the apparent diffusion coefficient increases with bed height ( 29 ; 35 ) , and theoretical estimates are frequently an order of magnitude larger than values obtained through inverse modeling. A key limitation of most analyses is the implicit assumption of constant porosity, despite the fact that even small variations in porosity or permeability can strongly influence pore-pressure dissipation rates. When grain-size distributions evolve, for example through clast comminution, both porosity and permeability may change significantly, making compaction and mixture compressibility central to the dynamics. In pyroclastic density currents, where fragmentation continuously generates fines and broadens the particle-size distribution, these effects can promote fragmentation-induced fluidization and sustained pore-pressure support ( 6 ) . Furthermore, if pore pressure dissipates through gas diffusion as described by Darcy’s law, the escaping gas necessarily reduces the gas volume fraction, implying that porosity cannot remain constant under such conditions. Capturing these coupled processes in predictive models remains a major challenge. High-fidelity multiphase simulations solve separate conservation equations for the solid and fluid phases, allowing spatial and temporal variations in pressure and volume fraction to emerge naturally. Such models reproduce fine-scale behaviours, including basal underpressures and overpressures beneath a propagating flow front, and the modulation of basal friction by dilatancy ( 7 ) . 1 used depth-resolved two-phase simulations of collapsing fluidized columns and showed that realistic dynamics could be obtained by adjusting an effective pore-pressure diffusion parameter consistent with static-bed measurements ( 29 ) . These simulations confirm an early high-mobility phase followed by a deposition phase as pore pressures dissipate, but also demonstrate that mis-specifying diffusivity can lead to substantial run-out errors. In parallel, reduced-order (depth-averaged) models have been developed to efficiently incorporate dilatancy effects. Building on the Coulomb mixture theory of 21 ; 17 , 18 introduced a depth-averaged framework that couples mass and momentum conservation with an evolving solid volume fraction and basal pore

Read the complete text