Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4546503 | Journal of Contaminant Hydrology | 2014 | 17 Pages |
•We mimic field-scale flow and solute transport in the laboratory using two columns.•Tracer breakthrough and temporal moments help conceptualize flow/solute transport.•A dual-porosity mobile–immobile approach simulates observed breakthrough curves.•Mass-transfer between mobile and immobile domains shows scale dependencies.
This study analyzed and compared unsaturated flow response and tracer breakthrough curves from a 10-m high constructed pile experiment (CPE) in the field (Antamina, Peru) and two 0.8 m high laboratory-based columns. Similar materials were used at both experimental scales, with the exception of a narrower grain size distribution range for the smaller column tests. Observed results indicate that flow and solute transport regimes between experimental scales were comparable and dominated by flow and solute migration through granular matrix materials. These results are supported by analogous breakthrough curves (normalized to cross-sectional area and flow path length) that suggest observation- or smaller-scale heterogeneities within the porous media have been homogenized or smoothed at the transport-scale, long breakthrough tails, and similar recovered tracer mass fractions (i.e., 0.72–0.80) at the end of the experiment. CPE breakthrough curves do indicate a portion of the fluid flow follows rapid flow paths (open void or film flow); however, this portion accounts for a minor (i.e., ~ 0.1%) component of the overall flow and transport regime. Flow-corrected temporal moment analysis was used to estimate flow and transport parameter values; however large temporal variations in flow indicate that this method is better suited for conceptualizing transport regimes. In addition, a dual-porosity mobile–immobile (MIM), rate-limited mass-transfer approach was able to simulate tracer breakthrough and the dominant transport regimes from both scales. Dispersivity values used in model simulations reflect a scale-dependency, whereby column values were approximately 2× smaller than those values applied in CPE simulations. The mass-transfer coefficient, for solute transport between mobile and immobile regions, was considered as a model calibration factor. Column experiments are characterized by a larger “mobile to immobile” porosity ratio and a shorter experimental duration and flow path, which supports larger mass-transfer coefficient values (relative to the CPE). These results demonstrate that laboratory-based experiments may be able to mimic flow regimes observed in the field; however, the requirement of scale-dependent dispersivities and mass-transfer coefficients indicates that these tests may be more limited in understanding larger-scale solute transport between regions of different mobility. Nevertheless, the results of this study suggest that the reasonably simplistic modeling approaches utilized in this study may be applied at other field sites to estimate parameters and conceptualize dominant transport processes through highly heterogeneous, unsaturated material.