Environmental Engineering Reference
In-Depth Information
extrapolate successful actions to additional species and habitats. h is will improve
costs of future control and restoration eff orts, while tracking performance meas-
ures and overall cost eff ectiveness.
Researchers and modellers must also track the accuracy and utility of modelling
capabilities for EDRA and document the economic and environmental savings by
using modelling products. Likewise, we must document customer satisfaction in
the use of modelling products to improve decision support.
Determining the spatial extent and severity of invasions is of utmost importance
(Simberloff
et al
. 2005). Unfortunately, ground surveys of each invasive species
require large amounts of time and funding, and most managers do not have the
resources required to complete the task. Statistical techniques linked to targeted
fi eld surveys may achieve fairly accurate measurements of potential distributions
in large areas. h ese models produce maps of habitat suitability or barriers to inva-
sion. h e information contained in remotely sensed images can be used in these
spatial models of habitat suitability (Reich
et al
. 1998, 2004;
Crosier 2004; Barnett
et al.
2007). h ese models provide information on the potential habitat of an
organism with minimal fi eld data on newly invading species. h ese methods could
prove invaluable for targeted early detection surveys.
While most would agree that reporting new locations of harmful invasive spe-
cies is important, there are a few published recommended data requirements
for early detection. Extreme minimum requirements include 'who, what, when,
and where' data (Table 3.2, required fi elds), sometimes referred to as the Dublin
GISINProc20041.html
and
http://dublincore.org
/
).
This general advice could be
greatly improved by an understanding of the potential to model species distribution
and abundance data in space and time. For example, ancillary data on abundance,
dominant native species present, other non-native species present, environmental
data (e.g. soils or disturbance information for plants, water depth for fi sh, nest tree
species for birds, etc.) and noticeably absent native and non-native species can be
extremely important information in predictive modelling (Table 3.2) (Morisette
et al
. 2006).
EDRA could be the most effective tools that land managers have to stop an inva-
sion before it becomes an ecological nightmare. A relatively modest investment
in existing global-scale information exchange systems will provide the world with
access to information about all known invaders. 'Watch lists' should be created,
maintained, and updated for local areas. When information is obtained about a
particular invasive species in a local area, it should be shared on global websites
and with local land managers so that others can benefi t from this knowledge. It is
