2003 Seattle Annual Meeting (November 2–5, 2003)

Paper No. 5
Presentation Time: 2:05 PM

A PRACTICAL APPROACH FOR SIMULATING SOLUTE TRANSPORT IN FRACTURED TILL


HELMKE, Martin F., Versar, Inc, 6850 Versar Center, Springfield, VA 22151, SIMPKINS, William W., Dept. of Geological and Atmospheric Sciences, Iowa State Univ, Ames, IA 50011 and HORTON, Robert, Agronomy Department, Iowa State Univ, Ames, IA 50011, helmkmar@versar.com

Hydrogeologists frequently encounter the problem of simulating solute transport through fractured till in glaciated terrains.  Despite strong evidence that fractures control solute transport in till, and the availability of modeling tools, relatively few studies apply models to the fractured till problem.  We hypothesize that the avoidance of such models results from either the perceived burden of obtaining input parameters or a lack of knowledge about these models.

In this study we demonstrate how transport parameters for fractured till may be obtained, and how they may be used to construct solute transport models.  Eight till units in Iowa (from depths of 1 to 27.5 m) and three conservative tracers (Br-, PFBA, and PIPES) were evaluated.  Critical parameters included bulk hydraulic conductivity (Kb) calculated by Darcy’s Law; fracture spacing (2B) measured in the field; fracture aperture (2b) estimated using the Cubic Law; and the effective diffusion coefficient (De) estimated using radial diffusion cells.  Values of Kb ranged from 7.7 x 10-10 to 3.8 x 10-5 m/sec; 2B ranged from 3.8 to 17.4 cm; 2b ranged from 4.0 x 10-6 to 1.1 x 10-4 m; and mean values of De were 5.6 x 10-10 (Br-), 2.9 x 10-10 (PFBA), and 1.3 x 10-10 m2/sec (PIPES).

The input parameters were used to construct three classes of models – the Mobile-Immobile Model (MIM), the Parallel-plate Discrete Fracture Model (PDFM), and the 3-Dimensional Discrete Fracture Model (3-D DFM).  The results from the models were then compared to breakthrough curves (BTCs) generated in the laboratory.  Model predictions of Br- were not significantly different between the models (a=0.01); however, the 3-D DFM was more accurate than the MIM or PDFM when predicting PFBA and PIPES BTCs.  For longer timescales and transport distances typical of field-scale investigations, the MIM is likely to be accurate, efficient, and easy to construct.  To demonstrate this, a transient, 2-dimensional version of the MIM was expanded to the field-scale using MODFLOW and MT3DMS.  This study demonstrates that the input parameters and computer tools available to the hydrogeologist make simulation of solute transport through fractured till both practical and informative.