Geoscience Reference
In-Depth Information
marker evaluation studies. Evaluation studies are a critical step in the marker
validation process, but many commonly used markers have been evaluated for
only a few species. Our review highlighted several specific deficiencies in our
understanding of marker effects, such as effects of radiotransmitters in fish,
efficacy of internally versus externally mounted transmitters (particularly
stomach implants), influence of external tags on amphibians and mammals,
and effects of transmitters on large mammals, particularly carnivores. It will
not be possible to evaluate the effects of all markers in all species, but results of
evaluation studies will be more useful if researchers focus their initial evalua-
tion studies on broadly applicable markers and species that share several bio-
logical features with other species of interest. We think that PIT tags will
become very important for marking animals in the future, and thus urge
researchers to undertake the necessary evaluation studies without delay. All
evaluation studies should be designed to measure marker effects across ages
and sexes.
Evaluation studies must include comparison to unmarked (or alternatively
marked) controls and adequate sample sizes. Statistical power analysis can be
used to calculate the minimum sample size required to detect significant
effects. Captive or laboratory conditions should be used whenever possible,
although researchers should ensure that biological parameters being measured
are not significantly affected by captivity (Berteaux et al. 1996). Ultimately,
results of captive studies should be validated in the field (Wilson and Culik
1992; Wallace et al. 1994; Cohen 1994). Also, evaluation studies should take
advantage of sensitive techniques for estimating subtle negative effects, such as
biomechanical modeling (Pennycuick 1989) and doubly labeled water (Ges-
saman and Nagy 1988). Marker evaluation studies should attempt to address
indirect effects of markers on survival (e.g., predation, starvation, disease) and
behavior (e.g., altered foraging or parental abandonment of juveniles to com-
pensate for higher energy demands), and focus not only on short-term but also
on long-term effects of markers (Reed and Oring 1993; Buehler et al. 1995;
Meyers et al. 1996).
Marker evaluation studies should be published in the peer-reviewed liter-
ature. It was not possible for us to determine whether a publication bias existed
in our sample of marker evaluation studies, but we suspect a tendency exists
for publishing studies that show significant negative effects over those failing
to show effects. Such a bias can misrepresent the effect of markers, and it in-
hibits dissemination of information about the best marking methods. Clearly,
some journals must encourage authors to perform and publish marker evalua-
tion studies. A similar suggestion was proposed to overcome the apparent pub-
