Ecology Theory

Extinction and population viability analysis

November 3, 2011, 1:34 pm
Content Cover Image

Artist reconstruction of an extinct Ammonite. Source: Nobu Tamura

Extinction and population viability analysis are inherently linked, since a minimum viable population is required by definition to sustain a species in the wild. Species extinction began an accelerating trend in the early Holocene concurrent with man's ascension as a dominant species of planet Earth.  This extinction trend continues to the present time as a result of anthropogenic disturbances in nature; this chain of extinctions is often referred to as the Holocene Mass Extinction. 

caption An endangered Painted hunting dog, Botswana
@ C.Michael Hogan

Although species extinction is a natural part of the biological world, the background extinction rate is deduced to be approximately one species per million per annum.E.O Wilson  estimates the number of extinctions per annum to be approximately 27,000 species, with a further estimate that up to half of all rainforest species may be extinct by the year 2030 at present rates of rainforest destruction. (Wilson) Ehrlich et al estimate that between 1600 and 1975 AD the extinction rate rose to fifty times the background level due to anthropogenic factors. (Ehrlich et al.) The US Fish and Wildlife Service Endangered Species Program has listed 1358 endangered plant and animal species in the United States; however, this number is not meant to be an accurate number of threatened species, but a governmental sanctioned lower bound, biased by an interest in large and easily observable species. Extinction is accompanied by a reduction in biological diversity. Concern over species extinction and biodiversity necessarily begs the question, “Why is preserving biodiversity important?” Organisms do not exist in isolation in the natural environment but instead are tied together in a complex web of interactions. Removal of a single taxon from this web can have widespread negative and sometimes unforeseen ecosystem impacts. Additionally, ecosystems with reduced biodiversity are  less able to provide goods and services to humans.

Extinction processes

In order to address biological diversity it is first necessary to understand the extinction process.  The seemingly simple question “How and why do species go extinct?” is actually quite complex. A population is technically considered extinct when the population size reaches zero. However, populations may be effectively extinct  prior to reaching a zero population. Once a population drops below a critical size it becomes extremely difficult for it to recover and it is functionally, extinct. This critical size is uniquely defined for each population of interest. Generally, multiple factors interact to cause a population to decline below their threshold size. These factors include habitat loss, a reduction in food supply, disease outbreaks, random catastrophic events and human pressures.

When a population’s size is reduced below their threshold number, the population is said to enter an extinction vortex. As populations decline, multiple factors work together to force the population size even further downward. For example, genetic diversity tends to be lost as populations decline in size. Genetic drift has a stronger effect in small populations and random alleles are fixed or lost at a higher rate. Inbreeding is more common in small populations because mate choice is limited. Inbreeding further reduces genetic diversity by lowering the number of heterozygotes in a population. Genetic variation allows for the long term survival of populations in variable habitats because it is more likely that some individuals will have traits that are well suited to new conditions. In the absence of genetic variation, populations are less resilient to natural and anthropogenic change. In addition, population declines are often accompanied by fragmentation from neighboring habitats. This fragmentation cuts off gene flow between populations and further reduces genetic diversity. The minimum number of individuals required to escape this extinction vortex is known as the “minimum viable population size”.

Minimum viable population size

A minimum viable population size is the smallest value for which a population maintains a high probability (typically >95%) of long term survival. Definitions typically take the form of a population size with an x% probability of surviving y years where x and y are defined for the specific system being studied. These values will vary significantly across taxa. Minimum viable population estimates are an extremely useful tool for conservation biologists. These numbers enable scientists to identify which species face the greatest risk of extinction and  prioritize conservation strategies. The estimates can also serve as recovery goals for endangered species using the logic that once a population exceeds their minimum viable size, it has a high probability of persistence. Minimum viable population size estimates are typically generated through population viability analysis.

Population viability analysis is a term referring to several methods that use mathematical models to predict population dynamics. Several different forms of data may be incorporated into these projection models, which normally include measures of survival and fecundity for the population in question. These data are likely to vary by taxon life development so they may be either age- or stage-specific. Life history data are ideally gathered from the population being studied, but may also be compiled for published information on similar populations. Projection models can also incorporate information on habitat availability. Viability analysis may include a measure of genetic variation within a population if low genetic diversity is likely to play a role in population persistence. Depending on the population being studied, it may be incorrect to assume that these data will not change with time or population density. To account for this potential problem, projection models can be either variant or invariant with respect to time and density. Invariant models assume that the data do not change over time due to external factors. Variant models incorporate random variation and the potential for catastrophic events into estimates of future population sizes.

A given model predicts population size progression over time, often with age group definition. Virtually all of these models are stochastic in nature, so that they incorporate random variation into the model. Since the variables comprising future time states are not known with certainty, numerousl model runs (typically more than 1000) are made for the same intial conditions. Each model run uses a different set of future conditions that are randomly selected from a pool of conditions that are likely to occur. Models are run for a specified number of generations and the outcome yields the number of individuals at the end of the time period, and whether or not the population went extinct.  Each model run will have slightly different results because each run uses slightly different future conditions. From this set of results it is possible to determine the probability that the population will survive. To determine the minimum viable population size, this process is repeated several times, each with different starting conditions (i.e. different population starting size). Survival probabilities are calculated for each starting condition. The smallest starting population that still has a high probability of survival (typically >95 percent) is defined as the minimum viable population size. Since each run of the model has a slightly different output, it is also possible to calculate the error associated with the probability of survival. The minimum viable population size and the error associated with this estimate are then used to make conservation and management decision.

caption Threatened grizzly bear, Yellowstone NP. @ Sandra Dooley

History of population viability analysis

Population viability analysis was first popularized during the 1970s as a tool for recovering endangered populations. One of the first populations that this technique was applied to was the Yellowstone National Park grizzly bear population. Under the Endangered Species Act, this grizzly bear population was listed as endangered because their population was drastically reduced in size due to human pressures and habitat fragmentation. Using population viability analysis, scientists estimated that a minimum of 70-90 grizzly bears were required for the population to have a 95% chance of survival for 100 years. Due to protection from hunting and an increased amount of habitat, the Yellowstone grizzly bear population has increased in number throughout the past thirty years and the population has been delisted from the Endangered Species List. This example illustrates one way in which population viability analysis and minimum viable population size can be used to maintain biological diversity.

See also


  •  Edward O. Wilson. 1999. The diversity of life. 424 pages
  •  Paul R.Ehrlich, A.H. Ehrlich and J.P. Holdren. 1977. Ecoscience: Population, Resources, Environment, W.H. Freeman, San Francisco, Ca. 




(2011). Extinction and population viability analysis. Retrieved from


To add a comment, please Log In.