250
Prospects and challenges
4 development of more refi ned theoretical models,
e.g. of species – area effects (Whittaker et al. , 2005 ;
Richardson & Whittaker, 2010 ).
As discussed in these pages, much recent progress
has been made in these areas, but they are likely to
remain important research foci. We will briefl y discuss
each in turn.
First, scale dependency is of central importance in
diverse aspects of conservation biogeography. For
instance, our assessments of conservation baselines
are strongly infl uenced by the temporal frame of reference (Chapter 3 ), yet there remain many areas of the
world where we lack clearly resolved long - term perspectives on ecosystem dynamics (e.g. Birks, 2005 ;
Willis et al. , 2007 ). Similarly, assessment of the spatial
patterns of diversity, the location of hotspots and the
outcomes of strategic conservation planning exercises
have also been shown to exhibit scale - dependency (e.g.
Lennon et al. , 2001 ; Ara ú jo et al. , 2005a ). The effects
of anthropogenic infl uences generally, changes arising
from the introduction of non - native species (e.g. Olden,
2006 ; Foxcroft et al. , 2009 ) and the criteria applied to
assess the extinction risk assigned to plant or animal
species also are sensitive to scale parameters of the
system (e.g. Mart í n, 2009 ).
Second (as highlighted in Section 10.2.1 ), as more
and more genetic, taxonomic and distributional data
are becoming available for analysis, we need to develop
increasingly sophisticated means of determining
which components of diversity variation are artefacts
of collecting intensity or of analytical failings, as
opposed to the ‘ real ’ underlying biogeographical
pattern. In addition, we are in the midst of a phylogenetic revolution, fuelled by fast, cheap molecular DNA -
sequencing technologies, which promises not only to
continue to refi ne our knowledge of species identities
but also to continue to provide exciting advances in our
understanding of evolutionary relationships across
even large clades. This in turn opens up new possibilities and challenges in the use of evolutionary distinctiveness indices in conservation planning analyses (e.g.
Forest et al. , 2007 ; Cadotte & Davies, 2010 ).
Third, and a key repeated theme, is the need to
deploy sensitivity analyses to a wide range of issues in
modelling future processes and patterns of diversity
change – for example, in respect of the spread of alien
species (Gallien et al. , 2010 ; Smolik et al. , 2010 ); the
role of spatial autocorrelation in species ’ distribution
models (Veloz, 2009 ); and the forecasting of species ’
enormous benefi ts for our conceptual understanding
of biogeographical processes and relationships.
However, in the excitement generated by the online
availability of data sets that, until recently, would take
years of study to assemble, it is important not to lose
sight of the need for continued efforts to determine
taxonomies and phylogenies, to document community
ecology and auto - ecology, and to engage in continued
fi eld sampling to generate the rigorous, systematic and
unbiased data sets required for pure and applied biogeographical analysis. Indeed, one problem inherent in
the increasing availability of digital data sets is that it
becomes increasingly easy for scientists to undertake
sophisticated but poorly founded analyses of inadequately sampled systems with which they are insuffi -
ciently familiar, as they have not personally taken
years of painstaking efforts to assemble the data.
To guard against this prospect, it is important that
those building such databases make particular efforts
to record meta - data describing the properties of the
data sets, and that biogeographers undertake what we
may broadly term sensitivity analyses involving the use
of null or simulation models, analyses of the spatial
structure of their data, application of data quality fi ltering, etc., to tease out the real structure in their data
from the bias and noise that frequently exist (e.g. see
Hopkins 2007 ; Hortal et al. , 2007 ; Feeley & Silman,
2010 ). Moreover, high - quality biodiversity information systems and biogeographical analyses are not, in
themselves, the solution to global conservation problems. The hard decisions about where to invest resources
and what aspects of biodiversity and ecosystem function should be prioritized will still be dependent upon a
mixture of historical legacy, political necessity and
societal consent (see Chapters 2 and 3 ).
10.2.2 Improving m odels, s imulations
and f orecasts
We previously highlighted four generic themes which,
we argued, required concerted attention within conservation biogeography, these were:
1 scale dependency;
2 inadequacies in taxonomic and distributional data;
3 sensitivity analyses to develop improved understanding of the effects of model structure and parameterization (e.g. in relation to predicting future species
extinction rates); and
Prospects and challenges
4 development of more refi ned theoretical models,
e.g. of species – area effects (Whittaker et al. , 2005 ;
Richardson & Whittaker, 2010 ).
As discussed in these pages, much recent progress
has been made in these areas, but they are likely to
remain important research foci. We will briefl y discuss
each in turn.
First, scale dependency is of central importance in
diverse aspects of conservation biogeography. For
instance, our assessments of conservation baselines
are strongly infl uenced by the temporal frame of reference (Chapter 3 ), yet there remain many areas of the
world where we lack clearly resolved long - term perspectives on ecosystem dynamics (e.g. Birks, 2005 ;
Willis et al. , 2007 ). Similarly, assessment of the spatial
patterns of diversity, the location of hotspots and the
outcomes of strategic conservation planning exercises
have also been shown to exhibit scale - dependency (e.g.
Lennon et al. , 2001 ; Ara ú jo et al. , 2005a ). The effects
of anthropogenic infl uences generally, changes arising
from the introduction of non - native species (e.g. Olden,
2006 ; Foxcroft et al. , 2009 ) and the criteria applied to
assess the extinction risk assigned to plant or animal
species also are sensitive to scale parameters of the
system (e.g. Mart í n, 2009 ).
Second (as highlighted in Section 10.2.1 ), as more
and more genetic, taxonomic and distributional data
are becoming available for analysis, we need to develop
increasingly sophisticated means of determining
which components of diversity variation are artefacts
of collecting intensity or of analytical failings, as
opposed to the ‘ real ’ underlying biogeographical
pattern. In addition, we are in the midst of a phylogenetic revolution, fuelled by fast, cheap molecular DNA -
sequencing technologies, which promises not only to
continue to refi ne our knowledge of species identities
but also to continue to provide exciting advances in our
understanding of evolutionary relationships across
even large clades. This in turn opens up new possibilities and challenges in the use of evolutionary distinctiveness indices in conservation planning analyses (e.g.
Forest et al. , 2007 ; Cadotte & Davies, 2010 ).
Third, and a key repeated theme, is the need to
deploy sensitivity analyses to a wide range of issues in
modelling future processes and patterns of diversity
change – for example, in respect of the spread of alien
species (Gallien et al. , 2010 ; Smolik et al. , 2010 ); the
role of spatial autocorrelation in species ’ distribution
models (Veloz, 2009 ); and the forecasting of species ’
enormous benefi ts for our conceptual understanding
of biogeographical processes and relationships.
However, in the excitement generated by the online
availability of data sets that, until recently, would take
years of study to assemble, it is important not to lose
sight of the need for continued efforts to determine
taxonomies and phylogenies, to document community
ecology and auto - ecology, and to engage in continued
fi eld sampling to generate the rigorous, systematic and
unbiased data sets required for pure and applied biogeographical analysis. Indeed, one problem inherent in
the increasing availability of digital data sets is that it
becomes increasingly easy for scientists to undertake
sophisticated but poorly founded analyses of inadequately sampled systems with which they are insuffi -
ciently familiar, as they have not personally taken
years of painstaking efforts to assemble the data.
To guard against this prospect, it is important that
those building such databases make particular efforts
to record meta - data describing the properties of the
data sets, and that biogeographers undertake what we
may broadly term sensitivity analyses involving the use
of null or simulation models, analyses of the spatial
structure of their data, application of data quality fi ltering, etc., to tease out the real structure in their data
from the bias and noise that frequently exist (e.g. see
Hopkins 2007 ; Hortal et al. , 2007 ; Feeley & Silman,
2010 ). Moreover, high - quality biodiversity information systems and biogeographical analyses are not, in
themselves, the solution to global conservation problems. The hard decisions about where to invest resources
and what aspects of biodiversity and ecosystem function should be prioritized will still be dependent upon a
mixture of historical legacy, political necessity and
societal consent (see Chapters 2 and 3 ).
10.2.2 Improving m odels, s imulations
and f orecasts
We previously highlighted four generic themes which,
we argued, required concerted attention within conservation biogeography, these were:
1 scale dependency;
2 inadequacies in taxonomic and distributional data;
3 sensitivity analyses to develop improved understanding of the effects of model structure and parameterization (e.g. in relation to predicting future species
extinction rates); and