Do Good Fences Make Good Neighbors for Mesopredators?
Mesopredator release—the increase in abundance and/or density of small predators when large predator populations decline (Soulé et al. 1988)—is a core concept underlying my own PhD research in Kenya. Although there is solid evidence for mesopredator release effects in a wide variety of circumstances around the globe—from the increases in red fox populations following lynx declines in Sweden (Helldin et al. 2006) to the proliferation of cownose rays as a result of the decimation of shark populations in the Atlantic (Myers et al. 2007)—many questions about the mechanisms underlying this phenomenon remain unanswered.
Some researchers have questioned the purely correlative nature of the evidence for mesopredator release (Gehrt & Clark 2003). Is the pressure of top-down intraguild predation really strong enough to be the primary influence on the abundance and density of small predators? There are many potential nuances—resource availability, anthropogenic habitat disturbance, social structure, capacity for behavioral plasticity on the part of the small predator—that could influence mesopredator release responses, yet still have not been adequately explored.
One starting point for tackling this is to ask whether removal of a predator has the same effects on multiple mesopredators in a given assemblage. If not, what life history factors may contribute to the differences? A study published in the most recent edition of the Journal of Mammalogy investigates this precise issue by comparing the effects of predator removal on multiple smaller predators at the same locations (Kamler et al. 2013).
Africa is known for its incredibly diverse carnivore assemblages, and is a tempting place to test how multiple small predators respond to declines in larger predator. Kamler et al. chose to focus on a triad of canid mesopredators in South Africa: the black-backed jackal (Canis mesomelas; 6-12 kg), bat-eared fox (Otocyon megalotis; 3-5 kg, remember this guy from a recent post?), and the Cape fox (Vulpes chama; 2-4 kg). The researchers simultaneously sampled the canids on two properties: one with fencing that excludes jackals but allows the foxes free entry and exit, and one that allowed free flow of jackals as well as foxes.
The region’s larger predators, such as wild dogs (Lycaon pictus), spotted and brown hyenas (Crocuta crocuta and Hyaena brunnea, respectively), lions (Panthera leo), cheetahs (Acinonyx jubatus), and leopards (Panthera pardus) had all been extirpated on both properties. Although the jackal is a fairly diminutive predator, especially compared to many of its African neighbors, the extirpation of large carnivores in this area has put the jackal in the position of being the largest predator around. (See here for my post on Scientific American's 'Expeditions' blog that includes a discussion of why a mesopredator does not necessarily play the ecological role of an apex predator even if it is left as the largest-bodied predator in an assemblage).
Kamler et al. started with a very basic hypothesis: that the densities of both foxes would be higher on the jackal-free property than on the property that was open to jackals. Basic mesopredator release effect, right? But the researchers wanted to dig deeper. They posed a series of other predictions as well, based on knowledge about life histories of both fox species.
First, they predicted that both foxes would have smaller range sizes on the jackal-free property. This is because the foxes would no longer have to roam around in order to avoid and seek refuge from jackals, and could concentrate their activities on the highest quality habitat, reducing the space needed to sustain each family group. Similarly, it was predicted that both foxes would show different habitat selection patterns on the jackal-free property, since they could concern themselves primarily with resource availability rather than the relative risk from jackals in different vegetation types.
The next set of predictions took into consideration how differences in the natural history of the two fox species might affect their responses to jackal removal. Kamler et al. predicted that the group size of bat-eared foxes would decrease, but not group size of Cape foxes. They also predicted that Cape foxes would shift more of their activity to the daytime, becoming more diurnal in the absence of jackals, but that bat-eared foxes would not change their temporal activity patterns.
Finally, Kamler et al. predicted that there would be decreases in activity overlap between the two fox species when jackals were not present. This is because they would no longer be squeezed into microhabitats or other zones that provided lower predation risk, and would be free to spread out and stake out their own zones of the jackal-free property.
The team set out surveying the canids on both properties. For the foxes, they installed a radio telemetry collar on at least one member of each family group, so that each group could be located and observed in order to track population abundance and density. Telemetry data was also used to record movement patterns, which allowed estimation range size for each species. Scent stations and scat counts were used to track jackal densities. Habitat selection was determined by comparing the proportion of different habitat types that were available to the actual proportion they represented in the animals’ home ranges. In order to examine activity patterns, the researchers determined the proportion of moving telemetry detections of each animal in each of three temporal categories: day, sunset, and night. They also measured prey abundance for each species, to control for the potential effects of differential resource availability.
The researchers’ efforts yielded fascinating results, some expected and some unexpected. This study included quite a few predictions covering multiple species, so I’ll cover it with an itemized summary of the predictions and corresponding results.
Lower densities of both foxes on non-jackal property.
The Cape fox density was three times higher on the fenced property . . . but bat-eared fox density actually decreased by 37% in the absence of jackals, contrary to the prediction. The authors suggested that this was likely due to differences in termite abundance on the two properties, with the benefits of a jackal-free zone being overridden by the lower availability of the bat-eared fox’s favorite menu item.
Smaller range sizes for both foxes on non-jackal property.
Both the Cape and bat-eared foxes demonstrated significantly smaller home range sizes where there were no jackals, consistent with the prediction. It appears that when foxes don't have to worry about dodging jackals and can focus on finding the richest resource sites, they can afford to reduce the area they occupy on a regular basis.
Both foxes would select for significantly different habitat types on non-jackal property.
Neither fox showed significantly different habitat preferences in the absence of jackals. The bat-eared fox did show more of a tendency towards using bushveld habitat where jackals were absent (p = 0.069), possibly to take advantage of the termite berry plant that was more abundant than on the jackal-friendly property. Although not quite statistically significant, this does suggest that difference in food resources between the two sites were driving bat-eared fox habitat selection, not the presence or absence of jackals. Note that the definitions of the different habitat types were not based on resource abundance, so this is not necessarily inconsistent with the finding that foxes reduce their range sizes on the jackal-free property.
Smaller group sizes for bat-eared foxes on non-jackal property.
Bat-eared fox group size was 48% smaller on the non-jackal property. Group size in this species seems to serve a vigilance function against jackal predation, thus the foxes may not need to stick together in such large numbers when predation risk is significantly lowered.
Cape foxes would become more diurnal in the absence of jackals.
This was indeed the case, with 10-13% of telemetry signals of Cape foxes being recorded during the day time on the jackal-free property, compared to only 0-2% where jackals were abundant. The reason that bat-eared foxes did not show the same effect (and were not expected to) is that they primarily track their activity patterns to those of their prey, northern harvester termites. These insects are active at night for most of the year, creating a situation in which it is lucrative for bat-eared foxes to be nocturnal as well, despite the risk of jackal predation.
Decrease in spatial overlap of Cape and bat-eared foxes in the absence of jackals.
The core areas of the Cape and bat-eared foxes overlapped less on the jackal-free property, indicating that they were able to partition den sites and other resources more broadly when they didn’t need to worry about predation risk from jackals.
So, that is a lot to absorb, and now we are much more informed about the ways that jackal presence affects the lives of small foxes in South Africa. But what is the take-home message of this study?
A key implication is the the fact that the exclusion of jackals affected two similarly-sized foxes in very different ways on some parameters. As the authors put it, “several sublethal effects were species-specific, probably due to different evolutionary histories of the fox species and related constraints on behavioral plasticity." The 'good fences make good neighbors' premise seems to hold up for Cape foxes, which dramatically increased their densities and shifted their activity patterns when jackals were excluded. The bat-eared foxes, however, appear to be primarily limited by food resource. The bat-eareds actually showed a density response that was dramatically opposite that which would be predicted in a mesopredator release situation.
These results illustrate the fact that we cannot consider species interactions to be isolated from life history factors such as diet and social structure. Although there is still much evidence for mesopredator release effects in many systems, we cannot take for granted that small predators are primarily controlled by large ones. Foxes must avoid being eaten, but foxes must also eat. Future studies in other systems, including those with intact apex predator assemblages, will provide better insights into the conditions and possible resource “tipping points” that determine whether a small predator is controlled primarily by resource constraints or risk from larger predators.
To end with something comical, I recently had an interesting incident during my own data collection that involved a black-backed jackal. A big male jackal and a white-tailed mongoose (Ichneumia albicauda) showed up at a trap to investigate the bait at the same time.
The bigger guy doesn't always come out ahead!
Gehrt, S. D. & Clark, W. R. (2003). Raccoons, coyotes, and reflections on the mesopredator release hypothesis. Wildlife Society Bulletin, 31, 836-842.
Helldin, J. O., Liberg, O., & Gloersen, G. (2006). Lynx (Lynx lynx) killing red foxes (Vulpes vulpes) in boreal Sweden: Frequency and population effects. Journal of Zoology, 270, 657-663.
Kamler, J., Stenkewitz, U., & Macdonald, D. (2013). Lethal and sublethal effects of black-backed jackals on cape foxes and bat-eared foxes Journal of Mammalogy, 94 (2), 295-306 DOI: 10.1644/12-MAMM-A-122.1
Myers, R. A., Baum, J. K., Chepherd, T. D., Powers, S. P., & Peterson, C. H. (2007). Cascading effects of the loss of apex predatory sharks from a coastal ocean. Science, 315, 1846-1850.