16 – Why, what, how much, and is it worth it?^219Given there are usually large uncertainties about the ecological processes
underpinning threatened species persistence and determining the effectiveness of
management, it is generally advisable to monitor threatened species, threats and
environmental covariates that may mediate impacts (e.g. fire history, climate).
Which species to monitor?
Lindenmayer and Likens (2010) and Brazill-Boast (Chapter 17) advise that
monitoring a few species well is preferable to monitoring many species poorly. This
advice seems particularly pertinent in the case of threatened species. Omnibus
monitoring programs that seek to understand the state of multiple species
populations and how they vary with stressors and management inevitably use
sampling strategies that are suboptimal to detect change and effects for any
individual species (Chapter 20). The ability to detect change in one species is
traded against the number of species included in the monitoring program, even if
observation strategies (e.g. trapping, spotlighting, call–playback) are qualitatively
the same across species. When dealing with this trade-off, we tend to defer to rules
of thumb, such as identifying one focal species of interest, and then accepting that
the ability to detect change or the effects of management on other species will vary.
Box 16.3. Which species to monitor? Analysing candidate
indicator species for assessing the effectiveness of fox baiting in
Western AustraliaTulloch et al. (2011) provided a framework for prioritising species for monitoring in
order to report on the effectiveness of a fox control program in Western Australia.
Their approach ranked the value of an indicator species (Vi) according to:
Vi = LiNi(Pi – P 0 )/Ci
Where Li is the proportion of the landscape in which a species is detected, Ni is
the number of species belonging to the same guild as species i, Pi is the probability
that species i increases in population size in response to fox control, P 0 is the
probability that species i increases without fox control, and Ci is the cost of
monitoring. This shows that the value of an indicator species increases: as it occupies
more of the landscape (because it is easier to detect, so confirming changes in
abundance is easier); as the number of similar species increases (because it is a
surrogate for more species); as its response to fox control increases (again, making
changes in response to management easier to observe); and as the cost of monitoring
it decreases. Tulloch et al. (2011) used their framework to identify that the western
brush-tailed possum Trichosurus v. vulpecula was the most useful indicator of the
effectiveness of fox control – it is widespread and likely to respond to a reduction in
fox predation in a way similar to several other species in the region.