In this post, Associate Editor Johan du Toit discusses new Policy Direction “Slow intrinsic growth rate in forest elephants indicates recovery from poaching will require decades” by Andrea Turkalo, Peter Wrege, and George Wittemyer, published today.

Intrinsic population growth is related to body mass

The rate at which a population grows (r) under ideal conditions with no resource limitation, disease, or predation, is governed by the biology of the species as adapted to its environment.  It is thus ‘intrinsic’ to the species for which it is the theoretical maximum rate of population growth, and so in population ecology is called rmax.  Empirically verifying rmax is tricky because ideal conditions hardly ever occur.  Nevertheless, rmax is a population-level expression of the life-history traits of the species and for mammals we know those traits, which include age of reproductive maturity, inter-birth interval, longevity, etc., are all related to body mass.  And indeed, rmax itself can be predicted quite accurately for mammals using a regression model that simply requires the mean body mass of the species.  Larger mammals require longer periods of gestation and lactation and their offspring take longer to grow to the size at which they can begin reproducing, so rmax decreases allometrically with increasing body mass; the bigger the species the slower it breeds.  When this relationship is illustrated in textbooks it is typically the African elephant, as the largest living land mammal, that appears at the end of the line as the biggest and slowest breeder.  But it is now generally accepted by zoologists that there are actually two species of African elephant – the savanna elephant and the forest elephant – and the estimate of rmax for African elephants (around 6-7 % per annum) has only been empirically verified for the savanna elephant, which is the bigger of the two.

Elephants seem to breed slower in the forest than in the savanna

With forest elephants being about 80% of the size of savanna elephants it would be expected that, all other things being equal, their rmax should be somewhat higher than 7% per annum.  And, in the absence of better information, conservation practitioners currently assume it safe to use demographic rates of savanna elephants for modelling the effects of poaching or protection on forest elephants.  Now, however, Turkalo et al. have found that a forest elephant population grows at a much slower rate than expected.  Based on the intensive monitoring of a cohort of individually identified elephants visiting Dzanga Bai in southwestern Central African Republic during 1995-2009, Turkalo et al. were able to derive estimates of natality, juvenile survival, age of first reproduction in females, inter-calving interval, cohort size, mortality (separated into natural and poached), dispersal, and so on.  They came up with an observed population growth rate (r) of just 2.6% per annum after controlling for poaching and dispersal.  Dung counts around Dzanga Bai indicated that this population maintained a lower density than has been reported elsewhere in the forest elephant range, yet forest conditions were unchanged and so density-dependent factors could be ruled out.  Their inference – from this one well-monitored population – is that in fact rmax for forest elephants is substantially lower than that of savanna elephants, which is surprising because forest elephants are smaller.

Why, and so what?

Living in a primary rainforest in the tropics is not easy for a large herbivorous mammal restricted to feeding from the forest floor.  There is a huge abundance of vegetation, but most of the edible stuff is high in the canopy and so forest elephants graze in scattered forest clearings and browse on the foliage, bark and fruit of woody plants within their reach.  Fruit is highly nutritious but scarce and ephemeral, and woody plants produce a plethora of secondary compounds that can be toxic.  Forest elephants thus utilize an exceptionally high diversity of plant species – to avoid overdosing themselves on specific toxins while meeting their intake requirements – and depend on social learning to know what to eat and how to navigate between areas of fruit availability using a permanent network of forest trails.  Savanna elephants have it easy by comparison, mainly grazing and only having to depend on woody browse in the dry season.  This difference, suggest Turkalo et al., could explain why forest elephants breed at a much slower rate than expected; their intrinsic rate of population growth is geared to the nutritional conditions of their natural habitat.

Using an exponential population growth model, Turkalo et al. projected out their forest elephant population into various future scenarios and found that, if all human killing could be effectively and permanently stopped, it would take at least 30 years to reach twice its current size.  If human killing continued but at no higher level than had prevailed during the 15 years of data collection, the population would take almost 60 years to double.  That is three times longer than if forest elephants conformed to the ‘average’ savanna elephant population growth rate.  These sobering results have to reach the managers of wildlife populations and forest areas in tropical Africa where prevailing assumptions of sustainable offtake rates, whether legal or illegal, need urgent revision.  Forest elephants are not simply smaller versions of savanna elephants.  If the findings of Turkalo et al. are applicable to the species as a whole – and the ecological rationale for that is compelling – then forest elephant populations are especially vulnerable to humans with guns.  Because forest elephants depend on social knowledge to locate key resources in space and time, the loss of a matriarch is profoundly more significant than the numerical reduction of a population by one individual.  And because forest elephants depend on scattered clearings and fruit-rich patches of forest with permanent trails in-between, they can be ambushed at close quarters.  And then, even if the killing could somehow be stopped, a forest elephant population needs an inordinately long time to recover because, as it now appears, the rmax for this species is the lowest of all living land animals.