Ecological Risk Assessment of Diffuse and Local Soil Contamination Using Species Sensitivity Distributions - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 686

• This oldest use is validated in the sense that estimated safe levels (most

often HC5s based on SSD-NOECs) are usually associated with the absence of
biodiversity and functional impacts in field or semi-field studies.

• In essence, the oldest use is a method of ranking the relative toxicity of

compounds for soil organisms; and this relative ranking method appears robust.

• Furthermore, newer uses show that SSDs are a versatile, easy to use modeling

tool, with which a broad array of problems can be addressed in a wide array of
contexts, for many contaminants and contaminant mixtures, and for both struc-
tural (in regard to the presence of species) and functional (in regard to Ecosystem
Services) responses.

•

In this newer use, the SSDs rank contaminated sites in terms of estimated toxic
pressures of compounds or mixtures. The higher a local toxic pressure, the higher
the expected impact on species reared in such soils. There is evidence that this
way of ranking is robust too, despite the problem that validation studies also show
large influences of other stressors.

• The output of SSDs can be used directly, for environmental decision making or

in the format of soil quality standards, or as a basis to plan further data generation
or assessments.

• In addition to the scientific validation of compound toxicity and site ranking,

there is “validation by use”; when a method is formally used, practice could
deliver output that would challenge the acceptance of the method; regular use
of SSDs has, so far, not led to credible challenges to SSDs as regulatory tools.

• SSDs can be part of a tiered approach, and in that context they are among the

lower-tier methods; in such methods, they may help to focus the next step in the
assessment on certain sites, taxa or compounds.

• SSDs may be used in scenario analyses to explore the (cost) effectiveness

of alternative Risk Management scenarios; in this sense, various studies have
demonstrated the usefulness of such explorative studies.

• SSDs may predict THAT a certain magnitude of impacts is expected, e.g. on

“biodiversity” or “Ecosystem Services”, but not WHAT will exactly happen (like
“loss of species X”).

Acknowledgement The authors are indebted to Dick de Zwart, Miranda Mesman, Piet Otte,
Michiel Rutgers, Job Spijker, Frank Swartjes, Kees Versluijs and Arjen Wintersen for develop-
ing the contents of this chapter, and the editor, Hunter Anderson, and two other reviewers for their
stimulating comments and worthwhile suggestions on a previous draft. Although this chapter has
been technically reviewed and cleared by the U.S. EPA, it does not necessarily reflect U.S. EPA
policy.

References

AKWA (2001) Base document ten-years scenario sediments - focus on slugde (in Dutch). Advies
en kenniscentrum waterbodems, Utrecht, the Netherlands, 120p
Aldenberg T, Jaworska JS (1999) Bayesian statistical analysis of bimodality in species sensitivity
distributions. SETAC News 19(3):19-20