Extinction debt occurs because of time delays between impacts on a species, such as destruction of habitat, and the species' ultimate disappearance. For instance, long-lived trees may survive for many years even after reproduction of new trees has become impossible, and thus they may be committed to extinction. Technically, extinction debt generally refers to the number of species in an area likely to become extinct, rather than the prospects of any one species, but colloquially it refers to any occurrence of delayed extinction.
Extinction debt may be local or global, but most examples are local as these are easier to observe and model. It is most likely to be found in long-lived species and species with very specific habitat requirements (specialists). Extinction debt has important implications for conservation, as it implies that species may become extinct due to past habitat destruction, even if continued impacts cease, and that current reserves may not be sufficient to maintain the species that occupy them. Interventions such as habitat restoration may reverse extinction debt.
Immigration credit is the corollary to extinction debt. It refers to the number of species likely to immigrate to an area after an event such as the restoration of an ecosystem.
Terminology
The term extinction debt was first used in 1994 in a paper by David Tilman, Robert May, Clarence Lehman and Martin Nowak, although Jared Diamond used the term "relaxation time" to describe a similar phenomenon in 1972.
Extinction debt is also known by the terms dead clade walking and survival without recovery when referring to the species affected. The phrase "dead clade walking" was coined by David Jablonski as early as 2001 as a reference to Dead Man Walking, a film whose title is based on American prison slang
for a condemned prisoner's last walk to the execution chamber. "Dead
clade walking" has since appeared in other scientists' writings about
the aftermaths of mass extinctions.
In discussions of threats to biodiversity, extinction debt is analogous to the "climate commitment" in climate change, which states that inertia will cause the earth to continue to warm for centuries even if no more greenhouse gasses are emitted. Similarly, the current extinction may continue long after human impacts on species halt.
Causes
Jablonski recognized at least four patterns in the fossil record following mass extinctions:
- (1) survival without recovery
- also called “dead clade walking” – a group dwindling to extinction or relegation to precarious, minor ecological niches
- (2) continuity with setbacks
- patterns disturbed by the extinction event but soon continuing on the previous trajectory
- (3) unbroken continuity
- large-scale patterns continuing with little disruption
- (4) unbridled diversification
- an increase in diversity and species richness, as in the mammals following the end-Cretaceous extinction event
Extinction debt is caused by many of the same drivers as extinction. The most well-known drivers of extinction debt are habitat fragmentation and habitat destruction. These cause extinction debt by reducing the ability of species to persist via immigration to new habitats. Under equilibrium conditions, species may become extinct in one habitat patch, yet continues to survive because it can disperse
to other patches. However, as other patches have been destroyed or
rendered inaccessible due to fragmentation, this "insurance" effect is
reduced and the species may ultimately become extinct.
Pollution may also cause extinction debt by reducing a species' birth rate or increasing its death rate so that its population slowly declines. Extinction debts may be caused by invasive species or by climate change.
Extinction debt may also occur due to the loss of mutualist
species. In New Zealand, the local extinction of several species of
pollinating birds in 1870 has caused a long-term reduction in the
reproduction of the shrub species Rhabdothamnus solandri, which requires these birds to produce seeds. However, as the plant is slow-growing and long-lived, its populations persist.
Jablonski found that the extinction rate of marine invertebrates was significantly higher in the stage (major subdivision of an epoch
– typically 2–10 million years' duration) following a mass extinction
than in the stages preceding the mass extinction. His analysis focused
on marine molluscs since they constitute the most abundant group of fossils and are therefore the least likely to produce sampling errors. Jablonski suggested that two possible explanations deserved further study:
- Post-extinction physical environments differed from pre-extinction environments in ways which were disadvantageous to the "dead clades walking".
- Ecosystems that developed after recoveries from mass extinctions may have been less favorable for the "dead clades walking".
Time scale
The time to "payoff" of extinction debt can be very long. Islands that lost habitat at the end of the last ice age 10,000 years ago still appear to be losing species as a result. It has been shown that some bryozoans, a type of microscopic marine organism, became extinct due to the volcanic rise of the Isthmus of Panama. This event cut off the flow of nutrients from the Pacific Ocean to the Caribbean
3–4.5 million years ago. While bryozoan populations dropped severely at
this time, extinction of these species took another 1–2 million years.
Extinction debts incurred due to human actions have shorter
timescales. Local extinction of birds from rainforest fragmentation
occurs over years or decades, while plants in fragmented grasslands show debts lasting 50–100 years. Tree species in fragmented temperate forests have debts lasting 200 years or more.
Theoretical development
Origins in metapopulation models
Tilman et al. demonstrated that extinction debt could occur using a mathematical ecosystem model of species metapopulations. Metapopulations are multiple populations of a species that live in separate habitat patches
or islands but interact via immigration between the patches. In this
model, species persist via a balance between random local extinctions in
patches and colonization of new patches. Tilman et al.
used this model to predict that species would persist long after they
no longer had sufficient habitat to support them. When used to estimate
extinction debts of tropical tree species, the model predicted debts
lasting 50–400 years.
One of the assumptions underlying the original extinction debt model was a trade-off between species' competitive
ability and colonization ability. That is, a species that competes well
against other species, and is more likely to become dominant in an
area, is less likely to colonize new habitats due to evolutionary
trade-offs. One of the implications of this assumption is that better
competitors, which may even be more common than other species, are more
likely to become extinct than rarer, less competitive, better dispersing
species. This has been one of the more controversial components of the
model, as there is little evidence for this trade-off in many
ecosystems, and in many empirical studies dominant competitors were
least likely species to become extinct.
A later modification of the model showed that these trade-off
assumptions may be relaxed, but need to exist partially, in order for
the theory to work.
Development in other models
Further
theoretical work has shown that extinction debt can occur under many
different circumstances, driven by different mechanisms and under
different model assumptions. The original model predicted extinction
debt as a result of habitat destruction in a system of small, isolated
habitats such as islands. Later models showed that extinction debt could
occur in systems where habitat destruction occurs in small areas within
a large area of habitat, as in slash-and-burn agriculture in forests, and could also occur due to decreased growth of species from pollutants.
Predicted patterns of extinction debt differ between models, though.
For instance, habitat destruction resembling slash-and-burn agriculture
is thought to affect rare species rather than poor colonizers. Models
that incorporate stochasticity, or random fluctuation in populations, show extinction debt occurring over different time scales than classic models.
Most recently, extinction debts have been estimated through the use models derived from neutral theory.
Neutral theory has very different assumptions than the metapopulation
models described above. It predicts that the abundance and distribution
of species can be predicted entirely through random processes, without
considering the traits of individual species. As extinction debt arises
in models under such different assumptions, it is robust to different
kinds of models. Models derived from neutral theory have successfully
predicted extinction times for a number of bird species, but perform
poorly at both very small and very large spatial scales.
Mathematical models
have also shown that extinction debt will last longer if it occurs in
response to large habitat impacts (as the system will move farther from
equilibrium), and if species are long-lived. Also, species just below
their extinction threshold,
that is, just below the population level or habitat occupancy levels
required sustain their population, will have long-term extinction debts.
Finally, extinction debts are predicted to last longer in landscapes
with a few large patches of habitat, rather than many small ones.
Detection
Extinction
debt is difficult to detect and measure. Processes that drive
extinction debt are inherently slow and highly variable (noisy), and it
is difficult to locate or count the very small populations of
near-extinct species. Because of these issues, most measures of
extinction debt have a great deal of uncertainty.
Experimental evidence
Due
to the logistical and ethical difficulties of inciting extinction debt,
there are few studies of extinction debt in controlled experiments.
However, experiments microcosms of insects
living on moss habitats demonstrated that extinction debt occurs after
habitat destruction. In these experiments, it took 6–12 months for
species to die out following the destruction of habitat.
Observational methods
Long-term observation
Extinction
debts that reach equilibrium in relatively short time scales (years to
decades) can be observed via measuring the change in species numbers in
the time following an impact on habitat. For instance, in the Amazon rainforest, researchers have measured the rate at which bird species disappear after forest is cut down.
As even short-term extinction debts can take years to decades to reach
equilibrium, though, such studies take many years and good data are
rare.
Comparing the past and present
Most
studies of extinction debt compare species numbers with habitat
patterns from the past and habitat patterns in the present. If the
present populations of species are more closely related to past habitat
patterns than present, extinction debt is a likely explanation. The
magnitude of extinction debt (i.e., number of species likely to become
extinct) can not be estimated by this method.
If one has information on species populations from the past in
addition to the present, the magnitude of extinction debt can be
estimated. One can use the relationship between species and habitat from
the past to predict the number of species expected in the present. The
difference between this estimate and the actual number of species is the
extinction debt.
This method requires the assumption that in the past species and
their habitat were in equilibrium, which is often unknown. Also, a
common relationship used to equate habitat and species number is the species-area curve, but as the species-area curve arises from very different mechanisms than those in metapopulation based models, extinction debts measured in this way may not conform with metapopulation models' predictions.
The relationship between habitat and species number can also be
represented by much more complex models that simulate the behavior of
many species independently.
Comparing impacted and pristine habitats
If
data on past species numbers or habitat are not available, species debt
can also be estimated by comparing two different habitats: one which is
mostly intact, and another which has had areas cleared and is smaller
and more fragmented. One can then measure the relationship of species
with the condition of habitat in the intact habitat, and, assuming this
represents equilibrium, use it to predict the number of species in the
cleared habitat. If this prediction is lower than the actual number of
species in the cleared habitat, then the difference represents
extinction debt. This method requires many of the same assumptions as methods comparing the past and present.
Examples
Grasslands
Studies
of European grasslands show evidence of extinction debt through both
comparisons with the past and between present-day systems with different
levels of human impacts. The species diversity of grasslands in Sweden appears to be a remnant of more connected landscapes present 50 to 100 years ago. In alvar grasslands in Estonia that have lost area since the 1930s, 17–70% of species are estimated to be committed to extinction. However, studies of similar grasslands in Belgium, where similar impacts have occurred, show no evidence of extinction debt. This may be due to differences in the scale of measurement or the level of specialization of grass species.
Forests
Forests in Vlaams-Brabant, Belgium, show evidence of extinction debt remaining from deforestation that occurred between 1775 and 1900. Detailed modeling of species behavior, based on similar forests in England
that did not experience deforestation, showed that long-lived and
slow-growing species were more common than equilibrium models would
predict, indicating that their presence was due to lingering extinction
debt.
In Sweden, some species of lichens show an extinction debt in fragments of ancient forest. However, species of lichens that are habitat generalists, rather than specialists, do not.
Insects
Extinction debt has been found among species of butterflies living in the grasslands on Saaremaa and Muhu
– islands off the western coast of Estonia. Butterfly species
distributions on these islands are better explained by the habitat in
the past than current habitats.
On the islands of the Azores Archipelago, more than 95% of native forests have been destroyed in the past 600 years. As a result, more than half of arthropods on these islands are believed to be committed to extinction, with many islands likely to lose more than 90% of species.
Vertebrates
80–90% of extinction from past deforestation in the Amazon
has yet to occur, based on modeling based on species-area
relationships. Local extinctions of approximately 6 species are expected
in each 2500 km2 region by 2050 due to past deforestation. Birds in the Amazon rain forest continued to become extinct locally for 12 years following logging that broke up contiguous
forest into smaller fragments. The extinction rate slowed, however, as
forest regrew in the spaces in between habitat fragments.
Countries in Africa are estimated to have, on average, a local extinction debt of 30% for forest-dwelling primates. That is, they are expected to have 30% of their forest primate species to become extinct in the future due to loss of forest habitat. The time scale for these extinctions has not been estimated.
Based on historical species-area relationships, Hungary currently has approximately nine more species of raptors than are thought to be able to be supported by current nature reserves.
Applications to conservation
The existence of extinction debt in many different ecosystems has important implications for conservation.
It implies that in the absence of further habitat destruction or
other environmental impacts, many species are still likely to become
extinct. Protection of existing habitats may not be sufficient to protect species from extinction. However, the long time scales of extinction debt may allow for habitat restoration in order to prevent extinction, as occurred in the slowing of extinction in Amazon forest birds above. In another example, it has been found that grizzly bears in very small reserves in the Rocky Mountains are likely to become extinct, but this finding allows the modification of reserve networks to better support their populations.
The extinction debt concept may require revision of the value of
land for species conservation, as the number of species currently
present in a habitat may not be a good measure of the habitat's ability
to support species in the future.
As extinction debt may last longest near extinction thresholds, it may
be hardest to detect the threat of extinction for species that
conservation could benefit the most.
Economic analyses have shown that including extinction in
management decision-making process changes decision outcomes, as the
decision to destroy habitat changes conservation value in the future as
well as the present. It is estimated that in Costa Rica, ongoing extinction debt may cost between $88 million and $467 million.
In popular culture
- An episode of the CBS series Elementary was named "Dead Clade Walking", and featured a Nanotyrannus skeleton found "significantly above" the K-T boundary.
