Ecological niche models and the geography of biological invasions: a review and a novel application - Invasive Plants: Ecological and Agricultural Aspects - page 33

condition simplifies the analysis, Reverse-ENM can suggest, during any stage
of the invasion, whether the invader is occupying habitats in its invaded range
that it does not occupy in its native range.
Reverse-ENM, if applied across many different taxa and between many dif-
ferent regions, could facilitate a general assessment of the predictability of the
geography of biological invasions. Emergent patterns may highlight groups of
organisms most likely to follow the same ecological rules in their native and
invaded ranges. Fortunately or unfortunately, depending on the perspective,
ample study cases now exist involving numerous taxonomic groups and near-
ly every continent. The following section uses, as an example, the invasion by
the red imported fire ant into the United States; limitations to the broader
application of Reverse-ENM are discussed at the end of the chapter.
A test case: the red imported fire ant in the US
Introduction
As an initial application of our Reverse-ENM method, we compared the actu-
al and predicted distributions of the red imported fire ant (RIFA) in South
America and the continental United States (US). RIFA, which is native to sub-
Amazonian South America, was introduced first into the US in the 1930s in
Mobile, Alabama [20], and has since spread throughout most of the southeast-
ern United States, as well as parts of Texas, Arizona, Nevada and California.
Infestations have been found recently as far north as Maryland and Delaware
[21]. In many areas, RIFA persists only in human-modified habitats (e.g., irri-
gated lawns in the southwestern US). Mortality of colonies by cold winter tem-
peratures is suggested to be the critical factor limiting colony survival [22],
and therefore the northern distributional limit. See [23] for a comprehensive
review of the natural history of RIFA and its invasion into the US.
RIFA is an ideal invader to evaluate Reverse-ENM because i) its adverse
ecological, economical and social impacts have made it one of the most well
studied invasive species, ii) its record of introduction and subsequent spread
are well documented, and iii) its present native and introduced distributions are
known. Consistent with the assumptions of Reverse-ENM described above,
the success of RIFA as an invader has been attributed partially to release from
natural enemies [24]. In addition, the spread of RIFA in the US appears to be
reaching its equilibrium distribution, in that its range expansion has slowed
considerably, especially along its northern range limit [22].
Methods
We produced predictions of the distributional potential of RIFA using a desk-
top computer version of the Genetic Algorithm for Rule-set Production