Environmental Engineering Reference
In-Depth Information
survival probabilities, either directly (e.g. McCarthy and Parris 2004), or because
the marks make individuals more vulnerable to harvesters. In some cases it may be
possible to detect and control for such effects (Box 2.11), but they should ideally
be avoided altogether. Capture methods that seek to maximise returns, such as
baiting, should be used with caution because they are likely to yield unrepresenta-
tive samples if classes of animal respond differently to the bait. A final important
Box 2.11
Survival estimation in action: mark-recapture analysis of crayfish.
Box 2.4 illustrated closed population mark-recapture analysis in a study estimat-
ing crayfish abundance in Madagascar (Jones and Coulson 2006; Figure 2.11).
That analysis used data from a single five-visit survey. However, these surveys
were repeated four times over a 22-month period, allowing rates of survival
between surveys to be estimated. A range of models were fitted to the data, allow-
ing survival and capture probabilities to vary with size class, sex and time. The
number of times each crayfish was caught was also fitted as an individual covari-
ate in order to look for possible effects of capture on survivorship.
The most fully parameterised models fitted the data well, suggesting that any
heterogeneity in survival and capture probabilities was explained adequately by
the covariates used. The best fitting model showed generally lower survival rates
in smaller crayfish, and strong evidence for a negative effect of handling on sur-
vival in large crayfish. Controlling for this handling effect, large crayfish survival
rate was estimated at over 0.7, but the estimate was only around 0.4 if this effect
was ignored. This emphasises the need to take seriously the possibility that
research activity can have a strong effect on the parameters it attempts to estimate.
Fig. 2.11
The Malagasy crayfish
Astacoides granulimanus
, found in forest
streams, and harvested for food. They can also be affected by temporary
capture; repeated handling dramatically reduces their survival rate. Photo
© Julia Jones.
