No abstract

This paper reports on the use of pulsed-field gradient (PFG) magnetic resonance (MR) techniques to obtain displacement and velocity spectra of steady-state, saturated flow through a column packed with glass beads. The displacement spectra obtained by PFG MR correspond to travel-distance probability-density functions (PDF) for initial conditions of a concentration impulse in a column with zero concentration. These spectra show strong dispersion-time dependence, and depart from Gaussian-shaped PDFs for short dispersion times. These data provide estimates of the dispersion-time dependence of transverse and longitudinal dispersion coefficients. The longitudinal dispersion coefficient reaches its long-time behaviour more slowly than the transverse coefficient; long-time values obtained from MR data agree well with those calculated using existing empirical correlations. A model based on three components of apparent velocities and dispersion coefficients is sufficient to describe the time dependence of displacement spectra for water flow through the bead column. The short-distance component arise because of convection-dispersion-diffusion processes within the narrow necks between particles. The long-distance component, on the other hand, represents a macroscopic convection-dispersion process. This study shows that PFG MR flow spectroscopy is a simple but potentially useful method for the investigation of flow and hydrodynamic dispersion in porous media, especially for time-dependent phenomena.

[1] Recurrent ponded infiltration (RPI) was applied to two undisturbed samples (5.4-cm diameter, 9-cm height) of coarse sandy loam (Korkusova Hut', CR). The water distribution within the samples during RPI was monitored using magnetic resonance imaging (MRI); the soil internal structure was visualized by X-ray computed tomography (CT). Preferential flow and a decrease of the steady state infiltration rate between two successive infiltrations of RPI are typical for the soil studied. During the MRI-monitored experiment these phenomena were pronounced only in one sample, which facilitated their linking to specific features in the MRI results; the flow rate decrease was related to a reduction of the MRI-detected water content and a change of the spectrum of T 1 (a characteristic related to the water surface-to-volume ratio). The MRI methods employed could detect water in low-density regions and thereby captured potential preferential pathways; however, one-dimensional (1-D) MRI demonstrated that some results may be affected by fast flow.

Experiments on model and real soil blocks designed to assess the feasibility of using magnetic resonance imaging for three-dimensional mapping of the time-varying spatial distribution of water in structured soils are reported. The results show that, notwithstanding inherent problems in imaging natural soils with a significant iron content, experimental parameters can be identified which allow satisfactory images to be obtained. Magnetic resonance imaging may therefore provide important information on soil structure and water movement in dual porosity soils, with attendant benefits for the calibration of models of non-Darcian flow in such soils.

Magne c resonance imaging (MRI) was used to study the process of infi ltra on in an undisturbed core of coarse sandy loam during a repeated ponded infi ltra on experiment (RPI), in which the fi rst infi ltra on was conducted into a naturally dry sample and the second infi ltra on into a wet sample. The primary aim was to assess the change in distribu on of entrapped air in the undisturbed soil sample and its infl uence on steady-state fl ow rates. The RPI experiment was conducted on a sample 8.9 cm in diameter and 8.4 cm high. Concurrently with the MRI monitoring, pressure heads and fl uxes were measured. Mul pleslice maps of longitudinal relaxa on me (T1), which is a parameter related to the surface/ volume ra o, and proton density (M0), related to the water content, covered almost the en re volume of the sample. The T1 mapping was conducted during four stages of the experiment: steady-state fl ow of the fi rst (I1) and second (I2) infi ltra ons and the equilibrium stage a er drainage of the sample (Stages D1 and D2). During I1 and I2, the highest values of M0 and T1 were detected in the upper 3 cm of the sample, which is in good agreement with fi ndings of computer tomography showing lower sample density and higher porosity in this region. The drop in the quasi-steady-state fl ow rate that was observed between the I1 and I2 runs corresponded to the decline in T1 and M0 in the upper 4 cm of the sample. The decrease in T1 indicates replacement of water in large pores with trapped air, while a lower M0 signifi es a general decrease of water content in the aff ected areas.Abbrevia ons: a.u., arbitrary units; CT, computed tomography; D1 and D2, fi rst and second drainages, respec vely; I1 and I2, fi rst and second infi ltra ons, respec vely; MR, magne c resonance; MRI, magne c resonance imaging; RPI, repeated ponded infi ltra on; T1, longitudinal relaxa on me; T2, transverse relaxa on me; TE, me to echo; TR, repe on me.