Biology Reference
In-Depth Information
With many species it is difficult to detect a very small initial infestation in an
area the size of the United Kingdom and even more difficult when a larger area is
viewed. Additionally, for a taxonomic group like plants, a small initial infestation
like that described for the insect would be more easily eradicated, making the tem-
poral spread model unnecessary. This type of modeling effort is also quite data
intensive.
The leafminer model was possible because detailed temporal data from other
invaded countries existed that could be used to calibrate a spread model, and
detailed data for the initial infestation in the new location was also available. Even
the data collection methods relied heavily on prior knowledge of the species.
Generally, these data would not be available for a species unless the species was a
problem elsewhere or if it was a well-studied species in its native range, relegating
the usefulness of available statistical techniques to a limited number of species with
readily available data. In a paper describing a spatially explicit population dynam-
ics models for the spread of the grey squirrel (
Sciurus carolinensis
) in a region of
Italy, the authors acknowledge they were only successful because detailed data on
population parameters for the squirrel were already available from previous
research (Tattoni et al. 2006). Using these models, the authors were able to provide
guidance to management activities by applying different control and removal sce-
narios. Again, while these models may be very useful to managers, this type of
detailed population dynamic information will not be readily available for most
invasive species.
The above examples rely heavily on knowledge of the biology of the species. We
recently developed a very simple early warning method for the spread of plant spe-
cies in United States counties. One of the goals of the method was to develop a
technique that could generally be applied to invasive species as soon as they move
into an area, without requiring time to be spent collecting detailed information spe-
cific to that species. There are general climatic factors that affect the distribution of
all species, such as temperature and precipitation. However, these models do not
include a temporal component. They predict the potential distribution of a species,
but do not attempt to predict when a species will actually reach its maximum dis-
tribution. A combination of these types of methods may be necessary depending on
the availability of species specific data.
6.3.4 Remote Sensing
Another method to capture the spread of a species involves remotely-sensed
imagery. Here, the spectral signature of a species is captured at different locations
in the field and then, using one of several different algorithms, the signature is
used to try to identify the species in the image across its extent. Images contain-
ing spectral information from the reflectance of sunlight can be captured at the
same location through time, and the spread of species can then be determined by
examining sequential images from the same location. However, this methodology
