serious underestimation of the number of
“contaminated" sites (a number which will
vary directly with the number of criteria
set and inversely with the concentration
selected to represent contamination).
There was also an initial attempt to
remediate these sites to extremely strict
management criteria, although over time
the selection of more realistic goals
appears to be occurring.

Remediation Criteria

As stated in a review of environmental
agencies world wide, no agency had a
method to consist intly set generic soil
criteria. This isa comment onthe
lack of an adequate scientific basis

on which to set generic criteria and the
problems posed by using them. Do
Canadian authorities believe that they
discovered the science others missed; or
will they be the only ones willing to make
assumptions that other felt inadvisable?

Cniteria vs. Risk-Based Approach

The finest quality risk assessments used at
contaminated sites today use stochastic
methods and. provide risk managers with
the best available information on which to
base site decisions. The method proposed
by this subcommittee, for setting the
generic soil criteria, is also risk-based but
nowhere near as rigorous as a high
quality, site-specific risk assessment. The
subcommittee’s draft document proposes
that criteria be set using extremely
complex methads. Despite the complexity,
they are not scientifically defensible as
they are based largely on assumptions
‘rather than empirically verified findings.

Conservative Assumptions - The "Safe"
Choice?

The proposed criteria are based on very
“conservative" risk management criteria.
The selection of 1 in 10° as an acceptable
level of cancer risk is extremely difficult to
defend in the context of contaminated
‘Sites. First, with a background lifetime
risk of cancer in the total population of

in 3, a risk of 1 in 10° has absolutely no
real-world meaning in either individual

public health terms. Given the small
numbers of people “exposed" at these
sites, it should be clearly understood that
reducing risks from even 1 in 10‘ to 1 in
10° is unlikely to prevent even a
single case of illness. Whether
expenditures of large sums of
money on theoretical reductions in
risk without any real-world benefit is
worthwhile is a risk management decision
that needs to be addressed by senior
goverument officials, elected
representatives and members of the
public. [t should also be appreciated that
the lower the soil concentration criteria,
the greater the number of contaminated
sites and the larger the perceived
"problem" becomes.

Land Use: Residential,
Commercial/Industrial, Agricultural

Although setting different generic criteria
for these land uses may seem reasonable,
this approach is based on several
assumptions that are not realistic. One
difficuity is the diversity of uses at
agricultural, residential and commercial
sites. For example, it is difficult to :
conceive of one general exposure scenario
that applies across Canada at all
agricultural sites. Is it reasonable to apply
the same generic agricultural soil criteria
to land used for: 1) vegetable
farming in Holland
Marsh, 2) growing
apples i ig the Okanagan, 3) growing oil
seed in Alberta or, 4) grazing caitle in
Saskatchewan? Is it possible to develop a
single “scientifically defensible" soil
criterion to these vastly different —
agriculture scenarios? If the "problem" is
contaminated foodstuffs, setting tolerances
might be a more effective approach.

Other Approaches

1. It may be possible to come up with
scientifically defensible generic soil
quality criteria for more limited
ee ential. s than the current

ae tial, agricultural and
senvercial categories. For example,
C dren’ s playground grade could
rived using a r¢ ‘asonably