The distribution of diversity: challenges and applications
69
For conservation purposes, this somewhat theoretical discussion has considerable practical signifi cance.
First, scientists in areas of the world that have been
poorly surveyed for particular taxa may be interested
in modelling what the current distribution of species
may be, in order to guide further survey work and to
contribute to analyses of the possible failures of systematic conservation planning based upon the known
ranges of species. An example of this approach was
given earlier in this chapter (Figures 4.2 , 4.3 ). In that
example, Hopkins (2007) used the known distributions to project the possible unknown ranges of species,
and by overlaying maps for 1,584 species was able to
generate maps of possible species diversity and of sampling defi cit.
In general, the more solid data we have on the presence and absence of a species across the range, the
easier it should be to identify those environmental variables that best represent the factors controlling the
distribution, although for species of particularly small
ranges this becomes particularly diffi cult to be sure of.
This approach, because of its reliance on climate data,
is termed (bio)climatic envelope modelling (BEMs or
CEMs; Pearson & Dawson, 2003 ), or sometimes ecological niche modelling. It is a growing fi eld with an
increasing array of computer programs developed for
the purpose (see further discussion in Chapter 7 ).
BEMs seek to establish statistical relationships
between the environmental variables and distribution
of a species. Testing of models is generally undertaken
by reserving a portion (typically 30 per cent) of the
past history, and potential future of plant and animal
species. The geographical distribution of any given
species is constrained by the fundamental niche
requirements of the species. These often are evident in
relationships with climatic variables such as (for plants)
mean annual temperature, growing degree days,
annual rainfall, or the seasonality of water and energy
regimes, but they may also be refl ected in particular
edaphic or other habitat variables. Distributions are
constrained within this fundamental or potential niche
space by biotic interactions with competitor species,
pests, pathogens, predators and mutualists, such that
the realized niche is always a sub - space of the fundamental niche. In addition, distributions are constrained
by dispersal limitations and past history of the geographical connectivity of their niche space. We know
this to be so from fossil records that show the presence
of species at some past time in an area in which present
climate is suitable for their growth and persistence, but
from which the species has been lost at some point in
the past due to past climate change and from the
numerous cases whereby humans have introduced
species into areas outside their natural range, only for
them to become troublesome invasives (Chapter 9 ).
Hence, the natural geographical ranges of species are
constrained within their potential ranges by:
1 biotic interactions;
2 the vagaries of history;
3 linked with 2 , the diffi culties of dispersing viable
propagules to all areas possessing appropriate conditions for their establishment.
population estimates across a large sample of species for different contexts, e.g. mainland versus
island systems (Mart í n, 2009 ).
The consequences of insuffi cient data on geographical distributions (the Wallacean shortfall) for
conservation prioritization is illustrated by the case of the Tanimbar corella ( Cacatua goffi ni ), a parrot
species endemic to the Tanimbar islands (Indonesia), a group of 66 islands of a total land area of
about 5,400 km
2 . This species was initially listed as ‘ threatened ’ in 1989 on the basis of small global
range and concern that it was being traded at a possibly unsustainable rate. However, a fi eld survey
on the largest island, Yamdena (3,250 km
2
), subsequently produced a population estimate of 231,500
( ± 33,000) for that island alone, suggesting that the initial categorization had been unwarranted
(Jepson et al ., 2001a ). There may also be some room for debate over whether the Tanimbar corella
represents a good species or a sub - species.
This aside, that an island species with a global range one - tenth the size of 50,000 km
2 can have
a healthy population size seems good news for conservation. Of course, this may be a misleading
example; it must be balanced against the knowledge that, for example, about half of Hawaii ’ s indigenous bird species have gone extinct since human colonization, and that population data indicate
many island species to be seriously threatened (Whittaker & Fern á ndez - Palacios, 2007 ).
69
For conservation purposes, this somewhat theoretical discussion has considerable practical signifi cance.
First, scientists in areas of the world that have been
poorly surveyed for particular taxa may be interested
in modelling what the current distribution of species
may be, in order to guide further survey work and to
contribute to analyses of the possible failures of systematic conservation planning based upon the known
ranges of species. An example of this approach was
given earlier in this chapter (Figures 4.2 , 4.3 ). In that
example, Hopkins (2007) used the known distributions to project the possible unknown ranges of species,
and by overlaying maps for 1,584 species was able to
generate maps of possible species diversity and of sampling defi cit.
In general, the more solid data we have on the presence and absence of a species across the range, the
easier it should be to identify those environmental variables that best represent the factors controlling the
distribution, although for species of particularly small
ranges this becomes particularly diffi cult to be sure of.
This approach, because of its reliance on climate data,
is termed (bio)climatic envelope modelling (BEMs or
CEMs; Pearson & Dawson, 2003 ), or sometimes ecological niche modelling. It is a growing fi eld with an
increasing array of computer programs developed for
the purpose (see further discussion in Chapter 7 ).
BEMs seek to establish statistical relationships
between the environmental variables and distribution
of a species. Testing of models is generally undertaken
by reserving a portion (typically 30 per cent) of the
past history, and potential future of plant and animal
species. The geographical distribution of any given
species is constrained by the fundamental niche
requirements of the species. These often are evident in
relationships with climatic variables such as (for plants)
mean annual temperature, growing degree days,
annual rainfall, or the seasonality of water and energy
regimes, but they may also be refl ected in particular
edaphic or other habitat variables. Distributions are
constrained within this fundamental or potential niche
space by biotic interactions with competitor species,
pests, pathogens, predators and mutualists, such that
the realized niche is always a sub - space of the fundamental niche. In addition, distributions are constrained
by dispersal limitations and past history of the geographical connectivity of their niche space. We know
this to be so from fossil records that show the presence
of species at some past time in an area in which present
climate is suitable for their growth and persistence, but
from which the species has been lost at some point in
the past due to past climate change and from the
numerous cases whereby humans have introduced
species into areas outside their natural range, only for
them to become troublesome invasives (Chapter 9 ).
Hence, the natural geographical ranges of species are
constrained within their potential ranges by:
1 biotic interactions;
2 the vagaries of history;
3 linked with 2 , the diffi culties of dispersing viable
propagules to all areas possessing appropriate conditions for their establishment.
population estimates across a large sample of species for different contexts, e.g. mainland versus
island systems (Mart í n, 2009 ).
The consequences of insuffi cient data on geographical distributions (the Wallacean shortfall) for
conservation prioritization is illustrated by the case of the Tanimbar corella ( Cacatua goffi ni ), a parrot
species endemic to the Tanimbar islands (Indonesia), a group of 66 islands of a total land area of
about 5,400 km
2 . This species was initially listed as ‘ threatened ’ in 1989 on the basis of small global
range and concern that it was being traded at a possibly unsustainable rate. However, a fi eld survey
on the largest island, Yamdena (3,250 km
2
), subsequently produced a population estimate of 231,500
( ± 33,000) for that island alone, suggesting that the initial categorization had been unwarranted
(Jepson et al ., 2001a ). There may also be some room for debate over whether the Tanimbar corella
represents a good species or a sub - species.
This aside, that an island species with a global range one - tenth the size of 50,000 km
2 can have
a healthy population size seems good news for conservation. Of course, this may be a misleading
example; it must be balanced against the knowledge that, for example, about half of Hawaii ’ s indigenous bird species have gone extinct since human colonization, and that population data indicate
many island species to be seriously threatened (Whittaker & Fern á ndez - Palacios, 2007 ).