[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries

Life expectancy in animals

Have you ever wondered what the life expectancy in captivity is of the white-cheeked gibbon (Hylobates leucogenys)? Or of Geoffroy’s marmoset (Callithrix geoffroyi) or the spectacled bear (Tremarctos ornatus)? Have you ever asked yourself whether the life expectancy of animals in captivity depends on whether they were born there or in the wild, or if the difference in life expectancy between men and women is repeated in other animals? Maybe you have never thought of these questions as you toured around a zoo or safari park. However, a recent article in the journal Demographic Research has attempted to provide some answers to these questions, making use of a large dataset of information relating to captive animals. This journal, published by the Max Planck Institute for Demographic Research, is available free of charge online at www.demographic-research.org.
In summary, the paper’s authors, in addition to providing estimates of life expectancy for a wide range of animals, found that as in man, female survivorship generally exceeds that of males; the life expectancy in captivity of animals born in the wild is not that different from that of animals born in captivity and, while mortality in most species rises with age, in other species it seems to be fairly level.

A CMI for animals?
Tracking animals in the wild can be a very hard task and is often only possible for small numbers of animals, hardly enough to generate any credible statistics about mortality rates. In the case of captive animals there is an international dataset, maintained by the International Species Information System (ISIS) that captures vital events from some 650 zoos and aquaria in 70 countries. The dataset covers over two million individual specimens.
By life office standards the size of the ISIS dataset is small when broken down by species. In addition, the authors chose a smaller subset of around 35,000 individual animals from 51 species and captured vital events for these animals for the period 1 January 1998 to 31 December 2003. The results will therefore be subject to significant random error. To overcome this, at least to some extent, they also looked at life expectancy by groups of homogeneous animals. For example, they calculated the combined life expectancy for a group of ape species containing the gorilla, orang-utan, siamang, and the white-cheeked gibbon. What the ISIS data shares with the data of life offices is that, at times, it needs careful treatment and understanding. For example, dates of birth for animals born in the wild and now in captivity often have to be estimated based on behaviour and morphology. They are often assigned a birth date of 1 January in a particular year. Another potential source of error is that still-births are often recorded as live births. This will affect short-term mortality rates and life expectancy from birth calculations.
Because of a lack of older animals in the ISIS data, and concerns about the under-reporting of deaths, the authors completed the life table by fitting a Gompertz curve to the mortality rates at younger ages. In the case of apes and crocodilians, so few deaths were recorded at older ages that the authors decided to produce life expectancies based on an assumed age by which all individuals would be expected to die. In the case of gorillas, for example, the highest assumed attained age was set at 40 while for crocodilians it was 30. The problem with the apes is unlikely to be due to under-reporting of deaths as these animals are among the most popular and visible animals in zoos. This makes the results for apes and crocodilians less reliable than for the other groups of species.

Born to be wild
Based on the ISIS data, the authors found that there was not much difference between the life expectancy of captive animals that were born in the wild compared to those born in captivity. However, many of the wild-born animals included in the ISIS dataset had entered captivity well before the period of observation and so will have survived the effects of the trauma often associated with being taken into captivity.
It is still a matter of debate whether wild animals, ie those never captured, have higher or lower life expectancies than animals raised in captivity. While animals in captivity have the benefits of veterinary care, a lack of predators, and a regular supply of food, they may suffer higher levels of obesity, poor adaptation to captivity or the zoo’s climate, and from inbreeding, leading to higher perinatal mortality. Living in close quarters may also make the spread of infections easier, including between species. This is not to mention the risk of poisoning from the food little Johnny throws to the animal while the keeper is not looking!

Life expectancies
The life expectancy (in years) of different groups of animals was calculated and is shown across in table 1.
Raptors (containing the bald eagle and king vulture) show a curious development of life expectancy, diminishing very slowly with age. The ‘ape’ category, which contains some of the closest animals to man in many ways, has the highest life expectancy even after allowing for the capping of the highest achievable age.
The life expectancy for some individual species was calculated and is shown in table 2.

Females versus males
The authors also investigated whether there were any noticeable differences between male and female mortality rates for the groups of species. They found that in half of the groups (apes, hoofstock, crocodilians, and kangaroos) female mortality was 33% to 40% lower than male mortality. Differences by sex were smaller and statistically insignificant in the other groups of species.

And finally
Spare a thought for Darwin’s rhea: 98% of the specimens in the dataset died before reaching their first birthday!