Agriculture Reference
In-Depth Information
protection values. He therefore preferred a distribution-based sensitivity threshold approach (Posthuma
et al. 2002) as scientifically more rigorous. This approach assumes that the variability in sensitivities
to toxicants among species is a function of inherent differences in species physiology and environ-
mental exposure conditions. Therefore, when considering the role of earthworms in the context of
risk assessment, it should be realized that they can serve only as a single potential species in the range
of other useful species. However, once the position of a species in the distribution function (
Figure
16.2
)
is known, its value as a test or indicator species may be derived. Species a in Figure 16.2 is
clearly more sensitive, and it may be more useful rather than species b or c to set environmental
quality criteria, provided that the same situations are under consideration. The same holds true for
the opposite approach of risk assessment (to determine the potentially affected fraction, PAF).
Risks of ecological damage also depend on what happens to a toxicant after it reaches the soil.
Toxicity may be a misleading term, especially if the chemical is one that will degrade quickly or
become detoxified. The extent of the initial environmental damage may be such that speedy recovery
from the impact can minimize damage within a short period of time. The extent and duration of
toxic effects are therefore important characteristics for categorizing chemicals (Kokta and Rothert
1992), but criteria by which recovery of organismal populations and communities can be measured
are still needed. Regulatory schemes are operated in many countries to ensure the environmental
safety of chemicals, especially pesticides. These schemes rely on experimental data obtained under
both laboratory and field conditions. In some schemes, it is a requirement that the chemicals must
also be examined for potential effects on earthworms (Greig-Smith 1992).
The International Workshop on Ecotoxicology of Earthworms held in Sheffield in 1991 rec-
ommended that a general risk assessment should be carried out for each type of land on which a
product is used and should be flexible. Consistent criteria for placing chemicals into categories of
high, intermediate, low, and negligible risk to earthworms should be agreed on. Products should
be labeled to indicate the degree of risk for earthworms. Most risk-assessment studies using
earthworms have been on agricultural land or relatively small areas of contaminated land. In case
of more diffuse pollution or large-scale use of chemicals, it may, however, be more appropriate to
focus on larger spatial scales. This requires alternative approaches, for instance, including landscape,
island, or metapopulation theories; definition of minimum viable population sizes; or assessment
of rates of recovery and dispersal rates (Eijsackers,
Chapter 17
, this volume).
A risk assessment scheme adopted by Kokta and Rothert (1992) still provides a useful procedure
applicable to earthworms, but ecological risk assessment in a broader sense is concerned not only
with effects of toxicants on earthworms, but also with endeavors to extrapolate their effects to
complex ecosystems. Ecological risk assessment essentially requires an interdisciplinary approach,
and earthworm researchers should remain aware of useful developments and approaches in other
related ecotoxicological fields. The incorporation of biomarkers into risk assessment schemes
(
Figure 16.3
) may contribute in the future toward a more scientific basis for decision making.
Advances in expert systems and artificial intelligence capabilities will contribute in the future to
the coordination of models and data.
The use of mathematical models for forecasting ecological effects of chemicals (Bartell et al.
1992) is well established in the aquatic field but much less so for the soil environment. Bartell et
al. (1992) considered that, although the original contention was that direct extrapolation of labo-
ratory toxicity data to the field was ill advised because of the complexity of ecological systems,
results from bioassays will continue to generate useful information because the effects measured
in laboratory microcosms can be compared with results from mathematical models. Various sim-
ulation models have been proposed for the soil environment (Axelsen 1997; Van Wensem 1997).
By applying these to earthworm communities, researchers in earthworm ecology and ecotoxicology
can contribute to a better understanding of the basic functions of soil ecosystems and the effects
of chemicals on them. This will contribute to the development of more useful risk assessment
methodologies.
