skip to main content


Title: African wild dog movements show contrasting responses to long and short term risk of encountering lions: analysis using dynamic Brownian bridge movement models
Abstract Background Prey depletion is a threat to the world’s large carnivores, and is likely to affect subordinate competitors within the large carnivore guild disproportionately. African lions limit African wild dog populations through interference competition and intraguild predation. When lion density is reduced as a result of prey depletion, wild dogs are not competitively released, and their population density remains low. Research examining distributions has demonstrated spatial avoidance of lions by wild dogs, but the effects of lions on patterns of movement have not been tested. Movement is one of the most energetically costly activities for many species and is particularly costly for cursorial hunters like wild dogs. Therefore, testing how top-down, bottom-up, and anthropogenic variables affect movement patterns can provide insight into mechanisms that limit wild dogs (and other subordinate competitors) in resource-depleted ecosystems. Methods We measured movement rates using the motion variance from dynamic Brownian Bridge Movement Models (dBBMMs) fit to data from GPS-collared wild dogs, then used a generalized linear model to test for effects on movement of predation risk from lions, predictors of prey density, and anthropogenic and seasonal variables. Results Wild dogs proactively reduced movement in areas with high lion density, but reactively increased movement when lions were immediately nearby. Predictors of prey density had consistently weaker effects on movement than lions did, but movements were reduced in the wet season and when dependent offspring were present. Conclusion Wild dogs alter their patterns of movement in response to lions in ways that are likely to have important energetic consequences. Our results support the recent suggestion that competitive limitation of wild dogs by lions remains strong in ecosystems where lion and wild dog densities are both low as a result of anthropogenic prey depletion. Our results reinforce an emerging pattern that movements often show contrasting responses to long-term and short-term variation in predation risk.  more » « less
Award ID(s):
2032131
NSF-PAR ID:
10327134
Author(s) / Creator(s):
; ; ; ; ; ; ; ; ; ; ;
Date Published:
Journal Name:
Movement Ecology
Volume:
10
Issue:
1
ISSN:
2051-3933
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract

    Within a large carnivore guild, subordinate competitors (African wild dog,Lycaon pictus, and cheetah,Acinonyx jubatus) might reduce the limiting effects of dominant competitors (lion,Panthera leo, and spotted hyena,Crocuta crocuta) by avoiding them in space, in time, or through patterns of prey selection. Understanding how these competitors cope with one other can inform strategies for their conservation. We tested how mechanisms of niche partitioning promote coexistence by quantifying patterns of prey selection and the use of space and time by all members of the large carnivore guild within Liuwa Plain National Park in western Zambia. Lions and hyenas specialized on wildebeest, whereas wild dogs and cheetahs selected broader diets including smaller and less abundant prey. Spatially, cheetahs showed no detectable avoidance of areas heavily used by dominant competitors, but wild dogs avoided areas heavily used by lions. Temporally, the proportion of kills by lions and hyenas did not detectably differ across four time periods (day, crepuscular, early night, and late night), but wild dogs and especially cheetahs concentrated on time windows that avoided nighttime hunting by lions and hyenas. Our results provide new insight into the conditions under which partitioning may not allow for coexistence for one subordinate species, the African wild dog, while it does for cheetah. Because of differences in responses to dominant competitors, African wild dogs may be more prone to competitive exclusion (local extirpation), particularly in open, uniform ecosystems with simple (often wildebeest dominated) prey communities, where spatial avoidance is difficult.

     
    more » « less
  2. Introduction: Predators can affect prey not only by killing them, but also by causing them to alter their behavior, including patterns of habitat selection. Prey can reduce the risk of predation by moving to habitats where predators are less likely to detect them, less likely to attack, or less likely to succeed. The interaction of such responses to risk with other ecological processes remains relatively unstudied, but in some cases, changes in habitat use to avoid predation may be constrained by competition: larger, dominant competitors should respond freely to predation risk, but the responses of smaller, subordinate competitors may be constrained by the responses of dominant competitors. For large grazing herbivores, an alternative hypothesis proposes that smaller prey species are vulnerable to more predators, and thus should respond more strongly to predation risk. Methods: Here, we tested these two hypotheses with 775 observations of habitat selection by four species of obligate grazers (zebra, wildebeest, puku and oribi) in the immediate presence or absence of four large carnivores (lion, spotted hyena, African wild dog and cheetah) in three ecosystems (Greater Liuwa, Greater Kafue and Luangwa Valley). Patterns of predation within this set were described by observation of 1,105 kills. Results:Our results support the hypothesis that responses to predation risk are strongest for larger, dominant competitors. Even though zebras were killed least often, they showed the strongest shift into cover when carnivores were present. Wildebeest, puku and oribi showed weaker habitat shifts, even though they were more frequently killed. These patterns remained consistent in models that controlled for differences in the hunting mode of the predator (stalking, coursing, or intermediate) and for differences among ecosystems. There was no evidence that smaller species were subject to predation by a broader set of predators. Instead, smaller prey were killed often by smaller predators, and larger prey were killed often by larger predators. Discussion: Broadly, our results show that responses to predation risk interact with interspecific competition. Accounting for such interactions should help to explain the considerable variation in the strength of responses to predation risk that has been observed.

     
    more » « less
  3. Many African large carnivore populations are declining due to decline of the herbivore populations on which they depend. We recently noted that the densities of true apex carnivores like the lion and spotted hyena correlate strongly with prey density, but competitively subordinate carnivores like the African wild dog benefit from competitive release when density of apex carnivores is low, so the expected effect of a simultaneous decrease in resources and dominant competitors is not obvious. We found that when prey density drops below a tipping point, the relationship of wild dog density to prey density changes sign, and wild dog density declines. We also noted that ‘prey depletion provides a mechanistically direct explanation of patterns in wild dog dynamics that have been attributed to climate change’ (Creel et al., 2023). Woodroffe et al. concur that prey depletion is an important threat, but suggest that we fail to understand the logic of their assertion that “climate change is likely to cause population collapse” (Rabaiotti et al., 2022), because the “identification of climate change as a threat is not based upon observed temporal trends in wild dog demography”. This statement misses our fundamental point. The data that Woodroffe et al. analyzed were collected over a period with rising temperatures and declining prey populations, so whether or not one tests for a time trend in demography, the data themselves are affected by two patterns: 
    more » « less
  4. null (Ed.)
    Abstract Large carnivores have experienced considerable range contraction, increasing the importance of movement across human-altered landscapes between small, isolated populations. African wild dogs ( Lycaon pictus ) are exceptionally wide-ranging, and recolonization is an important element of their persistence at broad scales. The competition-movement-connection hypothesis suggests that adaptations to move through areas that are unfavorable due to dominant competitors might promote the ability of subordinate competitors (like wild dogs) to move through areas that are unfavorable due to humans. Here, we used hidden Markov models to test how wild dog movements were affected by the Human Footprint Index in areas inside and outside of South Luangwa National Park. Movements were faster and more directed when outside the National Park, but slowed where the human footprint was stronger. Our results can be directly and quantitatively applied to connectivity planning, and we use them to identify ways to better understand differences between species in recent loss of connectivity. 
    more » « less
  5. Abstract

    Spatiotemporal variation in predation risk arises from interactions between landscape heterogeneity, predator densities and predator hunting mode, generating landscapes of fear for prey species that can have important effects on prey behaviour and ecosystem dynamics.

    As widespread apex predators, humans present a significant source of risk for hunted animal populations. Spatiotemporal patterns of risk from hunters can overlap or contrast with patterns of risk from other predators. Human infrastructure can also reshape spatial patterns of risk by facilitating or impeding hunter or predator movement, or deterring predators that are themselves wary of humans.

    We examined how anthropogenic and natural landscape features interact with hunting modes of rifle hunters and mountain lionsPuma concolorto generate spatiotemporal patterns of risk for their primary prey. We explored the implications of human‐modified landscapes of fear for Columbian black‐tailed deerOdocoileus hemionus columbianusin Mendocino County, California. We used historical harvest records, hunter GPS trackers and camera trap records of mountain lions to model patterns of risk for deer. We then used camera traps to examine deer spatial and temporal activity patterns in response to this variation in risk.

    Hunters and mountain lions exhibited distinct, contrasting patterns of spatiotemporal activity. Risk from rifle hunters, who rely on long lines of sight, was highest in open grasslands and near roads and was confined to the daytime. Risk from mountain lions, an ambush predator, was highest in dense shrubland habitat, farther from developed areas, and during the night and crepuscular periods. Areas of human settlement provided a refuge from both hunters and mountain lions. We found no evidence that deer avoided risk in space at the scale of our observations, but deer adjusted their temporal activity patterns to reduce the risk of encounters with humans and mountain lions in areas of higher risk.

    Our study demonstrates that interactions between human infrastructure, habitat cover and predator hunting mode can result in distinct spatial patterns of predation risk from hunters and other predators that may lead to trade‐offs for prey species. However, distinct diel activity patterns of predators may create vacant hunting domains that reduce costly trade‐offs for prey. Our study highlights the importance of temporal partitioning as a mechanism of predation risk avoidance.

     
    more » « less