Introduction to Contaminated Site Management - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 52

useful for short-term risk appraisal. Measurements of cadmium in urine, however,
represent a measure of cumulative lifelong exposure. Since this represents the site
exposure history, is it defensible to assume this measure also is representative for
long-term future exposure, when no changes in exposure conditions are expected.
And with regard to supportive measurements, the representative time span of clay
content (decades to centuries) and pH (1-10 year), for example, largely differ.
1.7.4 Laboratory Data Versus Field Data
Appropriate Risk Assessment is all about what is happening in the real world, that is,
at contaminated sites. However, Risk Assessment would hardly be possible without
the support of laboratory experiments. These experiments are used for at least three
different purposes. First, for ethical and technical reasons, direct human toxicolog-
ical effect data are rarely available. These data need to be derived from laboratory
experiments with animals, following strict guidelines with regard to laboratory ani-
mal welfare. Also ecological effect data heavily depend on laboratory experiments,
since it is difficult to investigate effects on one specific species in the field. Second,
it is inconvenient to control standard environmental conditions in the field.
As a consequence, Human Health and Ecological Risk Assessment both strongly
depend on experimental laboratory data. To transfer the laboratory effect data
to human health, and ecological effect data and to real world applications (field
conditions), the laboratory effect data are divided by assessment factors (often
called extrapolation factors). These assessment factors, which are often in the
range of 10 to values as high as 10,000, cover intraspecies and interspecies dif-
ferences and, last but not least, uncertainties associated with laboratory studies,
which always cover a limited amount of tests, for a limited time span. A specific
case of intraspecies differences in Human Health Risk Assessment is the differ-
ing sensitivity within the human population. In this case, an assessment factor
could be applied, when politically feasible, to also protect the most sensitive part
of the human population. In case of effects data for a whole ecosystem, instead
of a single species, uncertainties due to interspecies variation are dominant. In
fact, a major problem in Ecological Risk Assessment is the extrapolation of obser-
vations from individual and population levels to the ecosystem level (Eijsackers
et al. 2008 ). Again, assessment factors can be used to cover the corresponding
uncertainties.
A second important use of laboratory studies relates to the assessment of input
parameters. These input parameters may include 'supportive parameters' such as
physico-chemical contaminant characteristics (vapour pressure, octanol-water parti-
tion coefficient (Kow), water-saturated permeation coefficients, etc.). But laboratory
measurements can also focus on 'basic' (more lumped) input parameters, such as
indoor air concentrations, or leachate concentrations. Often, such measurements are
prescribed in manuals, guidance documents or Decision Support Systems. Examples
of this are the measurement of the concentration in vegetables in Tier 3 of the
Dutch tiered approach used to determine the risk due to the vegetable consumption