DRC-2018-013571 - 0901a06880981517

June 8, 2018
Kathy Weinel
Quality Assurance Manager
Energy Fuels Resources (USA) Inc.
225 Union Boulevard, Suite 600
Lakewood, Colorado 80228

Dear Ms.Weinel:
Attached to this letter is a copy of Hydro Geo Chem’s September 25, 2001 report entitled
“Assessment of the Effectiveness of Using Existing Monitoring Wells for GWDP Detection
Monitoring at the White Mesa Uranium Mill, Blanding, Utah”. This report provided a quantitative
assessment using a numerical 3-dimensional groundwater flow and solute transport model
constructed using TRACRN, an integrated finite difference computer code developed at Los Alamos
National Laboratories. TRACRN has the capability of simulating gas and liquid flow, and solute
transport, within variably saturated porous media.
The model was constructed and parametrized based on Mill hydrogeologic information available at
that time. The model was designed to simulate hypothetical ‘point source’ leaks that could
potentially develop within cells comprising the existing tailings management system at the site and
that somehow bypassed the cell leak detection systems. The model was designed to be conservative
in that it assumed an average hydraulic conductivity for the perched zone that was relatively high
(6.6 x 10-5 centimeters per second [cm/s]) and a 10:1 ratio of horizontal to vertical hydraulic
permeability.
Based on data collected since 2001, these assumptions are likely too conservative and result in
underestimating the simulated performance of the tailings management system monitoring wells at
that time. Critical factors influencing the adequacy of the monitoring well spacing include 1) the
degree of lateral spreading of a ‘point source’ leak as the resulting seepage migrated downward
through the vadose zone before contacting perched groundwater, and 2) the rate of perched ground-
water flow beneath the tailings management system.
First, the greater the degree of lateral spreading within the vadose zone, the larger the footprint of the
seepage contacting groundwater, and the fewer wells needed to reliably detect the impact. Therefore,
the greater the degree of lateral spreading, the more widely spaced the monitoring wells can be.
Second, for a given point source (‘leak’) strength, the lower the rate of groundwater flow, the greater
the expected impact to groundwater. In addition, the lower the rate of groundwater flow, the more
likely that timely detection of the impact by the groundwater monitoring wells becomes because the
impacts are likely to spread out more in the lateral (cross-gradient) direction. Since the lines of
monitoring wells on the cell margins are oriented generally in the cross-gradient direction, the lower
the rate of groundwater flow, the larger the acceptable spacing between monitoring wells.
Kathy Weinel
June 8, 2018
Page 2

H:\718000\hydrpt2018\report\2001_modeling_report\modeling_cover_letter.doc
The modeling presented within the attached report likely overestimates groundwater flow and
underestimates the degree of lateral spreading. The geometric average hydraulic conductivity beneath
the tailings management system is likely at least 12% to 15% lower than as estimated in the model.
In addition, the geometric average hydraulic conductivity downgradient of the tailings management
system (approximately 1 x 10-5 cm/s) is 85% lower than the value used in the model. Thus the model
likely overestimated the rate of groundwater flow and underestimated cross-gradient spreading.
Furthermore, the ratio of horizontal to vertical permeability within the vadose zone is likely to be
significantly greater than the 10:1 used in the model. A ratio larger than 10:1 is consistent with the
sub-horizontal layering that is present in both the Dakota Sandstone and Burro Canyon Formations
underlying the tailings management system. The effective ratio of horizontal to vertical permeability,
and the degree of spreading of any potential future seepage, will also be enhanced by the presence of
interbedded sub-horizontal shale and/or coarse-grained (conglomeratic) horizons within the Dakota
Sandstone and Burro Canyon Formations, further reducing the number of wells needed for reliable
detection.
Please feel free to contact me if you have any questions or concerns.

Sincerely,
HYDRO GEO CHEM, INC.

Stewart J. Smith, R.G.
Associate Hydrogeologist

Attachment (1)