Agriculture Reference
In-Depth Information
Figure 1. A conceptual model illustrating the use of ecological niche models to investigate differences
in native and invasive niche use or occupancy. In Forward-ENM, occurrence points from the native
range are used to develop an ecological niche model (1). The niche model is projected into native geo-
graphic space, assessed for accuracy (2), and then projected onto another landscape to predict the
potential invaded range (3). Reverse-ENM employs an analogous process, differing in that occurrence
points from the
invaded
range are used to predict the potential
native
range. In the final step (4), the
native range predictions are compared and differences assessed. Bold, uppercase letters correspond to
the geographic predictions in Figure 2. Adapted from [8].
In this scenario, a traditional or Forward-ENM model will predict the invad-
er to be absent from areas where it has invaded (i.e., it will under-predict the
potential invaded range (omission error)). Conversely, if we build a niche
model from the invaded range and project it onto the native region, the model
should identify areas where the species does not occur in its native region. The
areas that the species does not occupy, but that are predicted to have the spe-
cies present, should represent portions of its ecological niche that are unex-
ploited in its native range. The evaluation of the differences between the for-
ward and reverse predictions of the native distribution is the basis for the
Reverse-ENM process and is a means for assessing differences in invaded and
native niche occupancy (step 4, Fig. 1).
Reverse-ENM is directly analogous to Forward-ENM, differing only in the
space from which occurrence points are taken and the “direction” of the mod-
eling process. The environmental data layers used as modeled niche dimen-
sions in Forward- and Reverse-ENM are identical. Although an equilibrium
Search WWH ::
Custom Search