Biology Reference
In-Depth Information
2.5.5 Predictive Models Applied to Multiple Sites
Predictive models can be applied to multiple sites. However, models developed at
one site may have poor prediction success at other sites, because the relative impor-
tance of different realized niche dimensions can change between areas (see above).
For this reason, multisite models should be based on information for species and
predictor variables from each site. If this is not possible, then predictions of inva-
sive species distributions in areas where the models could not be validated should
be interpreted very cautiously. It is also a strong argument for the need to validate
predictions in the field before full implementation of an early detection program.
2.5.6 Optimized Monitoring Plan
An optimized monitoring plan allows for a further reduction in the search area
required for early detection monitoring (Fig. 2.1) and an increase in efficiency
(Fig. 2.4). It integrates the results of a generalized monitoring plan, prioritized
monitoring plan, and predictive modeling (Fig. 2.2). After a generalized monitoring
plan is used to identify areas most susceptible to invasion, prioritization is employed
to narrow the search range within this area and to identify the species most impor-
tant to monitor for. Predictive modeling is then applied to these high priority spe-
cies to develop efficient monitoring plans for those species. In some instances, it
may make sense to first predict which species are most likely to be introduced to a
site or spread into areas of high conservation value. In either case, this would be the
most efficient use of resources at both the planning and implementation stages of a
monitoring program. These considerations are typically overlooked in most early
detection programs (M. Brooks pers. obs.).
The payoff from investing in the optimized monitoring plan would be in imple-
mentation. Obviously, it would result in a minimum area being targeted for moni-
toring (Fig. 2.1). However, it will also increase the probability that the species most
likely to be problematic and the sites where the species are most likely to occur
and/or have the most negative effects have been identified. This would help identify
the best type of monitoring and control efforts needed to reduce the likelihood of
colonization, spread, and impacts of those high priority species.
2.6 An Example of How to Apply the Monitoring Framework
The framework described in this chapter can be used to increase the overall efficiency
of early detection monitoring programs. With the addition of each successive moni-
toring approach, the extent of the area which is the focus of monitoring efforts can be
reduced (Fig. 2.1). An example of how this process can work is presented below for
a hypothetical management unit composed of typical landscape features (Fig. 2.5).
