INTRODUCTION


Purpose of this review

    The purpose of an environmental impact statement is to identify those
aspects of the environment that may be affected by a proposed project. In
the
case of the proposed mining operation at Crandon, Wisconsin, by Exxon
Corporation the potential impacts are broad. It is the purpose of this report
to review those which relate to the effects of the project on ground water
quantity and quality.

    After potential impacts have been identified, it is also the function
of
the environmental review process to approximate the magnitude of the impacts
and to ascertain whether mitigative actions are possible which would reduce
the impact to tolerable or even negligible levels. In Wisconsin the
environmental review process is used to protect the public from a violation
of
their right to a clean environment and from having thrust upon them the
responsibility of cleaning up messes made by others. In that light, I have
been retained by the Public Intervenor of the Wisconsin Department of Justice
to review all materials cogent to possible impacts on the ground water at
the
Crandon site.

    This report presents my findings as of May 30, 1986. It is based on the
May, 1986, DEIS made public by the Wisconsin DNR and on the materials upon
which that DEIS is based. It will concentrate on these questions:

    1.   Are all likely impacts on the ground water system at Crandon
         identified in the DEIS?

    2.   Are the maximum possible risks to the public associated with those
         impacts correctly identified?

    3.   Are the mitigation and contingency actions proposed in the DEIS
         adequate to deal with the maximum risks?

    4.   If the answer to 3 is no for a particular impact, then:

         a.   What part of the risk may have to be borne by the public?

         b.   What will be the consequences of improper mitigation of that
               impact?

Summary of review conclusions

    The DEIS identifies three major potential impacts of the Crandon Project
on the ground water system:

    1.   water level declines (drawdowns) in the vicinity of the mine caused
         by mine dewatering,



With the addition of some secondary impacts caused by proposed mitigation
actions, these represent a reasonable assessment of the spectrum of impacts
to
ground water which will occur.

    The method used to predict the magnitude of specific impacts relies very
heavily upon computer models. In general terms, the modeling process requires
first that existing ground water conditions at the site of interest be
measured as completely as possible. Then a model is constructed which
reproduces those observed ground water conditions as closely as possible.
Finally the model is perturbed (modified) to include conditions like those
expected when the project is underway. The results of these perturbed model
runs can then be compared to present-day conditions to ascertain the extent
of
the impact of the project on the ground water system. At Crandon, the DEIS
uses computer models to predict the magnitude of each of the three primary
impacts listed above.

    In more 'specific terms, a ground water model is a set of mathematical
equations written to approximate natural processes, be they ground water
flow
or lake seepage or contaminant transport. Solution of the equations requires
three site specific inputs:

    1.   Inclusion of physical parameters which represent the properties
of
         the material through which the water flows (these can include
         hydraulic conductivity, porosity and dispersivity, among others),

    2.   Mathematical definition of the boundaries of the ground water system
         (locations where ground water flow originates or stops), and

    3.   The starting (or initial) conditions of the phenomenon being
         modelled.

    Once these impacts have been defined, then they are used as constraints
on
the solution of the mathematical equations of the model, and that solution
is
the result (or output) of the model. Typically ground water models will be
designed to produce ground water levels or flow vectors or chemical
concentrations as output.

    It is important to note before any review of the ground water models
used at Crandon, that any model's results are only as good as the inputs
provided to it. The adage "garbage in, garbage out" is absolutely
true. In
addition, the model cannot produce meaningful results if the equations which
comprise it are wrong. Thus the results of a ground water model can be no
better than the weakest link among its three major inputs and its equations.

    Generally the inputs cannot all be measured as accurately or as frequently
as would be ideal for a model. Therefore a process known as calibration/
verification is used to test the model's validity. Real observations of water
levels or chemical concentrations are made at the site of interest and then
the model is run to simulate these existing conditions. Physical parameters
(such as hydraulic conductivity) are varied at sites where they haven't been