The issue of minimum population sizes for maintaining genetic variation to support the natural evolutionary process has been theoretically examined. In 1980, Ian Franklin proposed "rule-of-thumb" estimates of effective population sizes Ne of 50 and 500 for short-term and long-term conservation, respectively. The figure of Ne = 50 is the theoretical minimum needed to keep inbreeding at 1% per generation, a level deemed tolerable for domestic animals kept under benign conditions. Such a population, however, would continue to suffer a loss in genetic variation or heterozygosity with succeeding generations. Franklin suggested Ne = 500 for gain in genetic variation from mutation to balance loss through drift. These approaches have been criticized as being too vague or general for any application to real populations in the wild.
Interestingly, a review by Mark Boyce indicates that these figures are probably of appropriate magnitude for several vertebrate species, as seen from empirical studies. Several simulation studies of the larger mammals also show that Ne = 50-100 is needed for populations to be safe from extinction due to demographic and environmental stochasticity. Further development of theory based on limited empirical data suggests that much higher effective population sizes may be needed for maintaining genetic variation in the long term. While Michael Lynch and Russell Lande estimate that Ne of up to 5,000 may be needed, Ian Franklin and R. Frankham maintain that Ne on the order of 5001,000 is sufficient for maintaining variation.
A clarification has to be made here about the definition of effective population size. In genetic terms, the effective population size is the number of individuals that would be subject to the same amount of genetic drift if it were an ideal, constant population under an equal sex ratio of adults, random mating, and random distribution of offspring among parents in the population. Only adult or reproductive individuals are considered for computing Ne. Thus, Ne has to be distinguished from the total size of a population and is usually only a fraction of it. Among elephants, a skewing of the sex ratio in many populations, primarily because of ivory poaching, is possibly the most important variable that decreases the value of Ne in relation to the number of adults Ns in the population. Ne is computed as 4(Ne x Nf)/(Nm + Nf), where Nm and Nf are the number of breeding males and females, respectively.
What do the theory and the models tell us about minimum viable populations for elephants? Although an arbitrary distinction, the MVP needs to be considered from both the short-term and the long-term perspectives. The simulations I carried out using demographic data from Asian elephants indicated that a population of 65-80 individuals, with a potential (deterministic) growth rate of 0.5% per year and no real constraint on carrying capacity, would have a high (>99%) probability of surviving for 100 years (chapter 7). If adult male-to-female ratios are skewed beyond about 1 : 5, the corresponding population size needed would rise, reaching 120 at a 1 : 16 ratio when the carrying capacity is at least twice this value.
In any case, at these population sizes and sex ratios, the value of Ne would be much below 50. Such populations would continue to lose genetic variation. Therefore, I suggested that maintaining a total population of 100-300 elephants, depending on demography, sex ratio, and basic ecological pressures, should be the goal of managers to ensure the population's survival in the short term in the face of demographic and environmental stochasticity as well as genetic erosion. Populations smaller than these sizes are not necessarily doomed, but would require intensive management to prevent them being drawn into the extinction vortex.
With elephant populations, the main consideration for short-term conservation should be their persistence in the face of demographic and environmental stochasticity. Genetic considerations are secondary for more than one reason. First, the link between genetic variation and short-term fitness of a species population is still rather controversial. Second, the long generation time in elephants also means that any loss of variation will occur over only a few generations within a century. It is only when an elephant population is threatened with strong selection, such as from ivory poaching, that genetics could become an important consideration even in small populations.
Long-term persistence would need much larger population sizes. The Armbruster-Lande model that simulated African elephant populations in semiarid habitats subject to recurring drought cycles indicated a size of about 3,000 individuals to ensure a 99% probability of persisting for 1,000 years (chapter 7). Asian elephant habitats do not seem to be impacted by such severe drought occurrences or climatic variability. Although not explicitly modeled, population sizes smaller than 3,000 would presumably have an equally high chance of persisting over the long term. Therefore, a population of 1,000-3,000 elephants or more, depending on the continent and its climatic zone, may be targeted by managers for long-term conservation. If one were to factor in the more recent theoretical arguments mentioned in the discussion on maintaining genetic variation, about twice the above figures may be appropriate.
There are many African elephant populations larger than these numbers by an order of magnitude or more; an attempt should obviously be made to keep these as large as possible in genetic terms. With Asian elephants, one does not have the luxury of many large populations. There are fewer than 10 populations with over 1,000 elephants each, most of them in India. Even here, the effective population sizes of some are small; one example is Periyar in southern India, with a total population of over 1,000 elephants, but an Ne of only 20-30 due to the extremely skewed sex ratio. It is critical, therefore, that the few remaining large Asian populations be maintained with a long-term perspective.
Was this article helpful?