Environmental Engineering Reference
In-Depth Information
the model to the ecosystem or humans causes further uncertainties. The extrapola-
tion from three species of different trophic levels to the whole ecosystem is highly
burdened by extremely high uncertainties: the recommended uncertainty factor of
1,000 reflects the extent of uncertainty of extrapolation from acute toxicity tests.
Adverse effects of chemicals measured on rats bear high measurement error due
to the individual differences of rats, laboratory practices, the duration of the tests
(e.g., the rats certainly become aged by the end of a chronic test), and the expo-
sure scenario, e.g., chemical uptake by food. In addition, the differences between
humans and rats result in an extrapolation uncertainty, generally handled by using
an interspecies factor of 10. Another example for uncertainties can be found in
human exposure with regard to land uses and exposure scenarios: for example,
exposure to metals through eating fish shows high spatial differences. People living
in the Great Plain, Eastern Hungary, eat fish very rarely, but inhabitants of the
towns of Paks and Baja at the Danube and Szeged at the Tisza River or around
Lake Balaton may consume fish as priority food. Even so, fish-free eating habits
still can be found in fish eating areas, increasing the uncertainty in human exposure
modeling. Bioavailability of contaminants is also a factor which highly influences
uncertainties, and makes optimization in risk assessment difficult. Optimization
in risk assessment means combining minimum risk with the lowest uncertainty.
A scientifically sound argument for disclaiming the risk of a contaminant not
bioavailable at all should have low uncertainty, meaning that data should derive
from a dynamic and long-term (chronic) model, and simulating a realistic worst
case (see also Chapter 6 in Volume 2).
Uncertainties during the process of risk assessment and risk management of
hazardous chemicals are innumerable, Verdonck et al. (2007) call for improving uncer-
tainty analysis in the EU risk assessment procedure. In connection with the European
REACH regulation, a guide was produced on information requirements and chemical
safety assessment in Chapter 19 of Uncertainty Analysis (2008).
Environmental risk assessment of contaminated sites is burdened with some char-
acteristic uncertainties. ERM of the contaminated environment is closely related
to environmental monitoring. If monitoring data are uncertain, the basis of risk
assessment and the objective of risk reduction are not valid, their reliability is poor.
Heterogeneity of the real environment makes data acquisition and the creation of infor-
mation from these data difficult. In addition to extrapolation uncertainties, one faces
uncertainties due to interpolation between environmental data/measurement points,
the scale of which is greater than the scale of heterogeneity.
The type and scale of environmental heterogeneities vary within a wide range:
a microsensor can detect differences and heterogeneities in the soil or sediment in
micrometer or millimeter range, while geological and geochemical heterogeneities may
occur in centimeter, meter or kilometer ranges. Microorganisms cause heterogeneity in
the micrometer or nanometer scale. The sampling/measurement point distribution and
frequency should be fitted to the scale of heterogeneities. Characterization of an area
based on measured results relies on interpolation between the point data by taking
into account uncertainties.
In the following, we will list the main sources of uncertainties in ERM moving
along the contaminated site from task to task, regardless of whether the uncertainties
Search WWH ::




Custom Search