Sentinel Species - Ecological Biomarkers: Indicators of Ecotoxicological Effects - page 158

well known, allowing a good understanding of the fluctuations observed in the biomark-
ers measured in them (Michel et al. 1998).
However, several authors (Chapman 2002; McCarty et al. 2002; Galloway et al. 2004; Tabor
and Aguirre 2004) have expressed regret that the criteria of selection of these species do
not integrate enough of an ecological dimension. More recently, De Lange et al. (2010) pro-
posed the assessment of ecosystem vulnerability by using biological and ecological informa-
tion at different hierarchical levels, in order to bring more ecology into environmental risk
assessment, indeed into ecotoxicology, a view supported by Luoma and Rainbow (2010).
In 2001, Chapman showed the utility and relevance of aquatic oligochaetes in ecologi-
cal risk assessment, but he noted that, because their taxonomy is difficult and uncertain,
they have largely been neglected or mistakenly considered as representing only pollu-
tion tolerant taxa. Nevertheless, more recently, Tixier et al. (2011) indicated that now their
taxonomy and autecology are sufficiently known to allow identification to species level
and that, when considered at the species level, the oligochaete community represents a
highly diverse group with a wide range of ecological roles and a wide range of sensitivity
to pollution. Thus, they proposed an approach of sediment quality assessment based on
oligochaete community analysis to evaluate the ecological status of storm water ponds.
Furthermore, as pointed out by Lam and Gray (2001), it is not uncommon that in the
laboratory, the choice of the species to be tested falls on a particular genotype of the spe-
cies, whereas in the natural environment populations are made up from a mixture of geno-
types. These same authors also expressed doubts on the choice of species that are easy to
breed and manipulate in the laboratory, considering that such species are the very spe-
cies that are often intrinsically strongly tolerant to any type of stress, natural or resulting
from contamination. Athrey et al. (2007) tested the effect of such reduced genetic varia-
tion in the least killifish Heterandria formosa . They used laboratory populations from an
eight-generation experiment designed to study selection for resistance to Cd, compared
to a control laboratory population and to fish from the original source population. The
authors drew two conclusions from this study. The first one is that the loss of genetic
variation in genes not directly under selection means that populations that have managed
to adapt genetically to an environmental pollutant might be less able to adapt to other
environmental stressors. Athrey et al. (2007) underlined that this could be exacerbated by
negative effects on fitness-associated characteristics that are sometimes associated with
contaminant adaptation. The second conclusion is that the loss of genetic variation in their
laboratory lines appeared to be due to effective population sizes being much lower than
actual population sizes, pointing out the necessity of maintaining large laboratory popula-
tion sizes of organisms used in environmental toxicology studies.
7.1.2 Criteria for Selection of Sentinel Species
In order to reflect the intensity of contaminant exposures, to integrate the effects of the
various parameters acting in the environment, and to be able to determine the effects
of exposures that may be of weak intensity but endure over a long period, specimens of
sentinel species collected in the field must be representative of the local environment and
meet several conditions. However, in certain cases, specimens may also be transplanted
into the habitat of interest, and maintained at the site in enclosures [active biomonitoring
(ABM)] in order to follow any change (increase or decrease) in contaminant exposure, and
the associated impacts on the individuals, limiting any effects of interspecific biotic factors
that might affect interpretation.