This report was prepared u an account of work s p o n d by the United States Government. Neither the United States nor the kpanmant oi Energy, nor any of thdr employees, nor m y of thdr contractors, subcontractors, or W r employees, mrka any wamnty, expnu or Implied, or amumcr any legal liability or responsibility for the accuracy, completeness or usefulness of my information, apparatus, product or process d i w l d , or represents that in use would not infringe privately owned rights. The views, .opinions and condusions contained in this report am those of the contractor and do not necessarily represent those of the United States Government or the United States Department of Energy.

In recent years there has been a renewed interest in the recovery of crude oil from oil shale; however, there is much concern as to how to dispose of the huge quantities of spent shale that would be generated. The spent shales contain a large percentage of soluble salts and other constituents that may exceed permissible levels if they enter local surface and ground-water systems. The objective of this study was to demonstrate the capabilities of the UNSAT1D model for assessing this potential threat to the environment by understanding water movement through spent shale piles.Infiltration, redistribution, and drainage of water in a spent shale pile were simulated with the UNSAT1D model for two test cases: 1) an existing 35 m pile, and 2) a transient pile growing at a rate of 10 m/year for 5 years. The first test case simulated three different layering scenarios with each one being run for 1 year. The second test case simulated two different initial moisture contents in the pile with each simulation being run for 30 years. Grand Junction and Rifle, Colorado climatological data were used to provide precipitation and potential evapotranspiration for a wet (1979) and dry (1976) year, respectively. Hydraulic properties obtained from the literature on Paraho process spent shale [1.2 Mg/m 3 (75 lb/ft 3 ) and 1.5 Mg/m 3 (95 lb/ft 3 )], soil, and clay were used as model input parameters to describe water retention and hydraulic conductivity characteristics. Plant water uptake was not simulated in either test case. The two test cases only consider the evaporation component of evapotranspiration, thereby maximizing the amount of water infiltrating into the pile.The results of the two test cases demonstrated that the UNSAT1D model can adequately simulate flow in a spent shale pile for a variety of initial and boundary conditions, hydraulic properties, and pile configurations. The test cases provided a preliminary sensitivity analysis in which it was shown that the material hydraulic properties, material layering, and initial moisture content are the principal parameters influencing drainage from the base of a pile. Pile design, to minimize drainage, can be improved significantly by use of detailed hydrologic modeling similar to that described in this report.

Schematic Diagram of Consequence Analysis iv 4.1 Illustration of Proposed Logic for Discretization of Variations Encountered in a Large, Natural, Multilayered, Groundwater Regime. 4-3 4.2 All Nodes Based on Variation in Aquifer Thickness of Elevation of Different Hydrogeologic Units and Stream Locations 4-5 4.3 Illustrative Example of Describing Interfaces of Two Materials in Well Log at a Node Location 4-8 4.4 Subdivision of the Given Region Into Two-Dimensional, Mixed-Order, Finite Elements Based on Nodalization of Figure 4.2 4-9 I 4.5 Three-Dimensional Finite Elements Generated from Two-Dimensional Surface Elements and Well Log Details at Each Node 4-11 5.1 Illustrative Sample Plot from PLOTEL 5-2 5.2 The Output from Program GRIDIN Provides a Means for Three-Dimensional Display of the Input Data of Results. 5-5 5.3

Geologic and geohydrologic data for the Paradox Basin have been used to simulate movement of ground water and radioactive contaminants from a hypothetical nuclear reactor spent fuel repository after an assumed accidental release. The pathlines, travel times and velocity of the ground water from the repository to the discharge locale (river) were determined after the disruptive event by use of a two-dimensional finite difference hydrologic model. The concentration of radioactive contaminants in the ground water was calculated along a series of flow tubes by use of a one-dimensional mass transport model which takes into account convection, dispersion, contaminant/ media interactions and radioactive decay. For the hypothetical site location and specific parameters used in this demonstration, it is found that Iodine-129 (1-129) is the only isotope reaching the Colorado River in significant concentration. The 1-129 is transported to the river at a maximum concentration of about 1.84 x 10-8 microcuries per milliliter (~Ci/ml). This concentration occurs about 8.0 x 10 5 years after the repository has been breached. This 1-129 groundwater concentration is about 0.3 of the drinking water standard for uncontrolled use. The groundwater concentration would then be diluted by the Colorado River (mean flow of 7,000 ft 3 /sec to about 2.13 x 10-13 ~Ci/ml. None of the actinide elements reach more than half the distance from the repository to the Colorado River in the two-million year model run time. As an example, Radium-226, at a maximum concentration, has travelled 0.46 of the distance to the river in two million years. Radium-226 concentration in the ground water at that point in space and time is 1.02 x 10-7 ~Ci/ml. This exercise demonstrates that the W1SAP model system is applicable for analysis of contaminant transport. The results presented in this report, however, are valid only for one particular set of parameters. A complete sensitivity analysis must be performed to evaluate the range of effects from the release of contaminants from a breached repository. iii. . SUMMARY LIST OF FIGURES.

No abstract