Speaker
Description
Hydrodynamic transport in rough-walled geological fractures is governed by the strong spatial heterogeneity of the aperture field. Even in the purely advective limit, this heterogeneity produces pronounced velocity intermittency along streamlines, with fluid particles alternating between fast channelized regions and extended low-velocity or quasi-stagnant zones. Such intermittency generates broad residence-time distributions, breakthrough-curve (BTC) tailing, and nonlinear growth of plume spatial moments. We study these mechanisms using a Monte Carlo ensemble of synthetic self-affine fracture aperture fields with prescribed relative closure and correlation length. Depth-averaged Stokes flow is solved under the lubrication approximation, and advective transport is simulated through a time-domain random walk (TDRW) scheme that tracks particle trajectories and residence times. Across all realizations, the velocity distributions exhibit a robust excess of low velocities controlled primarily by the fracture closure, revealing the geometric origin of transport anomalies.
To upscale these dynamics, we represent the Lagrangian velocity series as a stochastic Ornstein–Uhlenbeck (OU) process, embedded within a one-dimensional continuous-time random walk (CTRW). This reduced model uses only the velocity distribution, the advective tortuosity, and an effective Lagrangian correlation length. Despite its simplicity, it reproduces the detailed simulations, including early-time ballistic spreading, late-time superdiffusive behaviour, and the characteristic power-law BTC tailing associated with intermittent advective transport.
This work clarifies the physical origin of anomalous purely advective dispersion in rough fractures and provides a predictive, computationally efficient framework for upscaling fracture-scale transport into broader subsurface flow models.
| Country | Italy |
|---|---|
| Acceptance of the Terms & Conditions | Click here to agree |
