Geoscience Reference
In-Depth Information
1.00
0.80
1.50
0.60
1.00
0.40
0.50
0.20
0.00
0.00
50
100
150
2
4
6
8
Predicted ecological distance
Evapim (mm)
2
0.4
1.6
0.2
0.0
-0.2
-0.4
-0.6
1.2
0.8
0.4
0
FIGURE 6.3
GDM relates the degree of biotic turnover with the associated environmental turnover. Map
panel represents a set of cells used to calibrate the model. The scatterplot is the model residuals with the
line representing the modelled species turnover. The lower left plot is the relative contribution of the top five
most important correlates of turnover. The upper right panel is the I-spline transformed function for the best
environmental correlate. The lower right panel is a boxplot of the model residuals. (From Burley, H.M. et al.,
Int. J. Geogr. Inf. Sci
., 26(11), 2065, 2012.)
Second, the GDM approach has been applied as a spatially global model. It is therefore affected
by the same issues of spatial scale, non-stationarity and anisotropy that affect all spatial models.
Initial research shows this is the case (Burley et al., 2012), but more research is needed.
Third, while an extremely complex process, it is also possible to infer the species composition
at unsampled locations (Mokany et al., 2011). This is a problem of high complexity due to the
enormous number of possible permutations that need to be assessed. This means there will be huge
uncertainties associated with such an approach, but it does have the potential benefit of informing
field surveys as to what species might be expected and thus what to look for.
6.3.3 r
eSerVe
d
eSign
Reserve systems are a critical component of the conservation of biodiversity. However, their design
is a computationally challenging process for which GC methods can be relevant.
Reserve boundaries are typically determined through a combination of conservation, economic
and political pressures, with expert opinion and qualitative assessments used to help prioritise sites
(Prendergast et al., 1999). Large reserve systems are almost always developed incrementally and
are often comprised of areas with different land use and tenure histories, and these increments have
historically been selected opportunistically (Pressey, 1994).
A key aim of reserve design is to minimise the set of sites included in the reserve, thus minimis-
ing the cost of land acquisition and maintenance, while also maximising the number of species
