Water Flow Monitored by Tracer Transport in Natural Porous Media Using Magnetic Resonance Imaging

Abstract

Magne c resonance imaging (MRI) was applied to the study of fl ow processes in model and natural soil cores. Flow veloci es in soils are mostly too slow to be monitored directly by MRI fl ow velocity imaging. Therefore, we used for the fi rst me diethylenetriaminepentaacetate in the form Gd-DTPA 2− as a tracer in spin echo mul slice imaging protocols with strong weigh ng of longitudinal relaxa on me T 1 for probing slow fl ow veloci es in soils. Apart from its chemical stability, the main advantage of Gd-DTPA 2− is the anionic net charge in neutral aqueous solu on. We showed that this property hinders adsorp on at soil mineral surfaces and therefore retarda on. We found that Gd-DTPA 2− is a very convenient conserva ve tracer for the inves ga on of fl ow processes in model and natural soil cores. With respect to the fl ow processes in the coaxial model soil column and the natural soil column, we observed totally diff erent fl ow pa erns. In the fi rst case, the tracer plume moved quite homogeneously in the inner highly conduc ve core only and the migra on into the outer fi ne material was very limited. A numerical forward simula on based on independently obtained parameters showed good agreement between experiment and simula on and thus proves the convenience of Gd-DTPA as a tracer in MRI for soil physical inves ga ons. The natural soil core, in contrast, showed a fl ow pa ern characterized by preferen al paths, avoiding dense regions and preferring loose structures. In the case of the simpler model column, the local fl ow veloci es were also calculated by applying a peak tracking algorithm.Abbrevia ons: DTPA, diethylenetriaminepentaacetate; MRI, magne c resonance imaging; NMR, nuclear magne c resonance; rf, radiofrequency; SEMS, spin-echo mul slice.Flow processes in soils control the water and nutrient supply in plants, contaminant transport, and the recharging of aquifers. Natural soils are heterogeneous media with locally varying grain size, pore size, and pore geometry. In addition, highly conductive structures like wormholes and decomposed roots are present. All these factors lead to so-called preferential pathways, along which water fl ows faster vertically than expected from the average fl ow velocity v av [(volume fl ow rate/area)/water content: v av = (Q/A)/θ = J w /θ] but from which water can also diff use laterally into neighboring small-pore-size regions. All these processes can strongly aff ect water redistribution in soils, its bioavailability for plants, and transport of contaminants Clothier et al., 2008). Classical investigations of such processes have been invasive and not temporally resolved. For example, soil cores are infi ltrated with dyes and aft erward sliced and photographed, so that only the static result of fl ow processes can be observed (Wehrhahn et al., 2007). Th e direct observation of such processes in situ is diffi cult due to the opaque nature of the systems. Th erefore, during the past two decades, tomographic methods have become popular for the investigation of soils and soil-r...

Transport of gadolinium- and arsenic-based pharmaceuticals in saturated soil under various redox conditions

“…Similar observations of Gd-DTPA migration with the tracer were reported by Haber-Pohlmeier et al (2010), who showed that Gd-DTPA behaves conservatively in sand and in a natural soil core. They reported no retention and explained that the Gd-DTPA is not adsorbed due to its large size.…”

Upward Transport in a Three‐Dimensional Heterogeneous Laboratory Soil under Evaporation Conditions

“…For the simulation of water flow, a three‐dimensional, cell‐centered, finite‐volume code was used (Ippisch et al, 2006; Rossi et al, 2008; Carminati and Flühler 2009; Haber‐Pohlmeier et al, 2010; Bechtold et al, 2011b) that solves the Richards equation: where h is pressure head (cm), u is the pore water velocity vector (cm d −1 ), K is hydraulic conductivity (cm d −1 ), and z is the vertical coordinate directed positively upward (cm). The variables θ( h ) and K (θ) were described by the van Genuchten–Mualem parametric expressions (van Genuchten 1980) (Table 1).…”

In Situ Magnetic Resonance Imaging of Plant Roots

“…Early MR studies already showed the effects of image distortion and signal loss in various types of soil (Bottomley et al, 1986). To minimize image distortions and in an effort to optimize MR relaxation parameters, sand was used as a medium for the latest MR studies (Haber‐Pohlmeier et al, 2010; Stingaciu et al, 2013). The researchers used a multislice, multi‐echo spin echo MR experiment.…”

“…The first studies focused on the ability to image roots, depending on soil conditions and an added MR contrast agent (Omasa et al, 1985; Bottomley et al, 1986; MacFall et al, 1991). Recent publications have reported the use of fast imaging techniques to reduce imaging time for the three‐dimensional acquisition of root images (Haber‐Pohlmeier et al, 2010) and the use of single‐point imaging (SPI) in systems with low water content (Gruwel et al, 2004). When the water content becomes low and the water MR signal starts to broaden, conventional MRI fails to produce an image while SPI can still provide useful images.…”