|
Recent improvements in radiotelemetry and remote-sensing
technology have greatly expanded the capability of researchers
to gather data on space and habitat use by animals. We can
apply sophisticated approaches to analysis of these data to
answer the classic questions concerning habitat selection and
home range size as well as to test for differences between
age/sex classes, times of day, seasons and large scale
treatments (e.g. logged vs. unlogged areas). Obtaining
meaningful answers to these questions requires careful choice
of the scale(s) of measurement (point location to homerange
area), predictor variables (continuous such as canopy
coverage or categorical such as vegetation types), sample
unit (location vs. animal) and sample size. The design must
avoid dependence of observations caused by pseudo-replication
of observations within animals, serial correlation through
time or spatial autocorrelation, or it must apply statistical
methods designed for such dependent observations. We
explored the implications of these improved radiotelemetry
and analysis tools for the design of current and future
radiotelemetry studies and suggest guidelines for estimating
necessary sample sizes.
|
|