Ecological Risk Assessment of Diffuse and Local Soil Contamination Using Species Sensitivity Distributions - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 659

construct the SSD. One should construct an SSD beyond the common one based
on NOECs when all output of an assessment would be obtained in the upper tail of
that SSD.
14.10.5 When Soil Concentrations in an Area Vary
For contaminated sites, it is common to see a huge variance in contaminant concen-
trations among soil samples. So, there are species sensitivity differences, but also
differences in exposure levels within a specific contaminated site. Hence, the out-
put of an SSD in terms of risk levels (PAF) in such cases would deliver a spatial
distribution of Potentially Affected Fractions . At some spots in the site (hot spots),
risks are extreme, while at other spots the risks are low. Similarly, a suite of data on
toxic risks may be collected in time. For example the risk reduction due to declining
concentrations as a consequence of breakdown of compounds or of leaching may
lead to a temporal distribution of PAFs.
For risk managers, it may be worthwhile to know the variability in toxic pressures
within a site or area, to help focus attention on hot spots (for remediation). A tempo-
ral distribution in PAF for an Outcome Assessment after taking Risk Management
actions may demonstrate that risks were indeed reduced as a consequence of those
actions.
Statistically, the need to combine the variability of exposures in space or time
with that of sensitivities (species) has lead to the derivation of Joint Probability
Curves (JPC). JPCs result from combining an exposure distribution with an SSD.
The results look like the graph in Fig. 14.12 , where a hypothetical site is sampled at
three moments in time, at multiple spots within a site: before and after remediation
of the most contaminated spots, and after autonomous breakdown of organic toxi-
cants over time. The graphs represent a Conventional Risk Assessment of the initial
situation (“high”), and two Outcome Assessments (results of the remediation “mod-
erate” and waiting “low”). Note that “an ideal remediation” would have resulted in
a sharp-angled JPC-curve, with the angle in the lower right. Such a curve reads as:
“there are only very few spots left with a toxic pressure of more than 5% of the
species”.
14.10.6 When There is a Mixture of Contaminants
When considering contaminated sites, it is obvious that the local biota are nearly
always exposed to mixtures of contaminants. Hence, a Risk Assessment should
address the issue of potential cumulative impacts. How to do that has been a problem
for a long time.
An often-applied approach in practice is simply not to model mixture responses,
and treat each compound separately. This is, in fact, an approach of “no addition”.
It is implicitly assumed that only the most toxic contaminant is causing the impacts.