58
Basic biogeography: estimating biodiversity and mapping nature
dinosaurs, fl ying reptiles, large sea reptiles and
ichthyosaurs.
The imperfect nature of the fossil record, the challenge of applying modern species concepts to fossilized
remains and the diffi culty of distinguishing mass
extinction events and pulses means that it is in turn
extremely diffi cult to estimate what might be thought
of as the ‘ business as usual ’ or background rate of
extinction. Those who have attempted such calculations estimate that most species persist for perhaps
around 4 million years, with a broad range of 1 – 10
million years average duration, allowing the background rate to be estimated based on average duration
and total richness in a group (Raup, 1991 ). By such
back - of - the - envelope calculations of the background
rate, anthropogenic extinction rates for birds and
mammals have been estimated to be 100 to 1,000
times faster than background (e.g. Primack, 1993 ).
Leaving apart the problems of calculating the background rate, where do the estimates of extinction rates
for the period of human - dominance of extinction
(sometimes termed the Anthropocene) come from and
how good are they?
The fi rst step in dealing with the question of estimating anthropogenic extinction rates, past and present, is
to recognize that the term ‘ extinction ’ is used in very
different ways, linked to differing sets of assumptions.
The typology of extinctions provided by Ladle & Jepson
(2008) provides a novel framework for exploring the
meaning of extinction (Table 4.2 ), in which the fi rst
two categories refer explicitly to the previous two sections of this chapter, viz . the Linnean and Wallacean
shortfalls. Hypothetical population trajectories that
might accompany these forms of extinction are provided in Figure 4.5 .
1 The term Linnean extinctions refers to attempts to
estimate extinctions for areas or regions that are poorly
known scientifi cally, but where the available data leads
scientists to believe that large numbers of species exist,
many of which are likely to be endemic to the region.
When this is applied, for example, to large swathes
of tropical forest in the equatorial regions, the use
of area - based extrapolations of how much diversity
may be present in the pristine state of these systems
allows us, in turn, to use species – area relationships to
predict how many species will be lost as the habitat is
destroyed.
In essence, the approach taken in such estimates is
closely akin to the Erwin method for extrapolating from
local sampling, based on rates of host specifi city and
immense, not least for convincing politicians and the
public of the need to take action. The crux of the
problem is that if we don ’ t know what is out there or
how widely species are distributed, how can we convince people about the reality and form of the biodiversity crisis? An equally problematic issue is how to go
about fi lling the shortfalls when funding for conservation in general – and taxonomy in particular – is
extremely limited.
4.2.3 The e xtinction e stimate s hortfall
Whereas the births and deaths of most individuals
are discrete, easily recognizable events, it is often diffi cult to determine when a new species has come into
existence, and when the last individual of a species
has died.
(Lomolino et al ., 2010 , p. 207)
Extinction is a natural process and it is often remarked
that of all the species that have lived, only a small
fraction are alive today. Processes generating species
extinctions over geological time periods include volcanic eruptions, meteorite impacts, climatic changes,
marine transgressions, ocean closures and the disappearance of lakes, in combination with biotic forcing
as new forms or newly arrived forms have displaced
others (Raup, 1991 ; Lomolino et al ., 2010 ).
Rates of extinction have varied throughout the
history of life (a period of some 3,500 million years),
so it is diffi cult to distinguish between so - called background rates of extinction and short episodes of unusually rapid or extensive losses. However, analysis of
the fossil record suggests that there have been fi ve mass
extinctions, each of which may be defi ned as a major
episode of extinction involving many different taxa
and occurring fairly suddenly in the fossil record. These
fi ve events are each recognizable in the marine record,
with the most recent three, the end - Permian, end -
Triassic and end - Cretaceous events also notable in the
terrestrial tetrapod (four - limbed vertebrates) record.
Other, lesser, pulses of extinction have also been recognized in the fossil record for terrestrial animals,
although mass extinctions are not clearly distinguishable for plants (Willis & Bennett, 1995 ; Willis &
McElwain, 2002 ). The most recent generally recognized mass extinction event, the so - called K – T event
(Cretaceous – Tertiary), occurred at around 65 million
years ago and saw the disappearance of the land
Basic biogeography: estimating biodiversity and mapping nature
dinosaurs, fl ying reptiles, large sea reptiles and
ichthyosaurs.
The imperfect nature of the fossil record, the challenge of applying modern species concepts to fossilized
remains and the diffi culty of distinguishing mass
extinction events and pulses means that it is in turn
extremely diffi cult to estimate what might be thought
of as the ‘ business as usual ’ or background rate of
extinction. Those who have attempted such calculations estimate that most species persist for perhaps
around 4 million years, with a broad range of 1 – 10
million years average duration, allowing the background rate to be estimated based on average duration
and total richness in a group (Raup, 1991 ). By such
back - of - the - envelope calculations of the background
rate, anthropogenic extinction rates for birds and
mammals have been estimated to be 100 to 1,000
times faster than background (e.g. Primack, 1993 ).
Leaving apart the problems of calculating the background rate, where do the estimates of extinction rates
for the period of human - dominance of extinction
(sometimes termed the Anthropocene) come from and
how good are they?
The fi rst step in dealing with the question of estimating anthropogenic extinction rates, past and present, is
to recognize that the term ‘ extinction ’ is used in very
different ways, linked to differing sets of assumptions.
The typology of extinctions provided by Ladle & Jepson
(2008) provides a novel framework for exploring the
meaning of extinction (Table 4.2 ), in which the fi rst
two categories refer explicitly to the previous two sections of this chapter, viz . the Linnean and Wallacean
shortfalls. Hypothetical population trajectories that
might accompany these forms of extinction are provided in Figure 4.5 .
1 The term Linnean extinctions refers to attempts to
estimate extinctions for areas or regions that are poorly
known scientifi cally, but where the available data leads
scientists to believe that large numbers of species exist,
many of which are likely to be endemic to the region.
When this is applied, for example, to large swathes
of tropical forest in the equatorial regions, the use
of area - based extrapolations of how much diversity
may be present in the pristine state of these systems
allows us, in turn, to use species – area relationships to
predict how many species will be lost as the habitat is
destroyed.
In essence, the approach taken in such estimates is
closely akin to the Erwin method for extrapolating from
local sampling, based on rates of host specifi city and
immense, not least for convincing politicians and the
public of the need to take action. The crux of the
problem is that if we don ’ t know what is out there or
how widely species are distributed, how can we convince people about the reality and form of the biodiversity crisis? An equally problematic issue is how to go
about fi lling the shortfalls when funding for conservation in general – and taxonomy in particular – is
extremely limited.
4.2.3 The e xtinction e stimate s hortfall
Whereas the births and deaths of most individuals
are discrete, easily recognizable events, it is often diffi cult to determine when a new species has come into
existence, and when the last individual of a species
has died.
(Lomolino et al ., 2010 , p. 207)
Extinction is a natural process and it is often remarked
that of all the species that have lived, only a small
fraction are alive today. Processes generating species
extinctions over geological time periods include volcanic eruptions, meteorite impacts, climatic changes,
marine transgressions, ocean closures and the disappearance of lakes, in combination with biotic forcing
as new forms or newly arrived forms have displaced
others (Raup, 1991 ; Lomolino et al ., 2010 ).
Rates of extinction have varied throughout the
history of life (a period of some 3,500 million years),
so it is diffi cult to distinguish between so - called background rates of extinction and short episodes of unusually rapid or extensive losses. However, analysis of
the fossil record suggests that there have been fi ve mass
extinctions, each of which may be defi ned as a major
episode of extinction involving many different taxa
and occurring fairly suddenly in the fossil record. These
fi ve events are each recognizable in the marine record,
with the most recent three, the end - Permian, end -
Triassic and end - Cretaceous events also notable in the
terrestrial tetrapod (four - limbed vertebrates) record.
Other, lesser, pulses of extinction have also been recognized in the fossil record for terrestrial animals,
although mass extinctions are not clearly distinguishable for plants (Willis & Bennett, 1995 ; Willis &
McElwain, 2002 ). The most recent generally recognized mass extinction event, the so - called K – T event
(Cretaceous – Tertiary), occurred at around 65 million
years ago and saw the disappearance of the land