Mixed‐species groups of Serengeti grazers: a test of the stress gradient hypothesis

Abstract

Understanding the role of species interactions within communities is a central focus of ecology. A key challenge is to understand variation in species interactions along environmental gradients. The stress gradient hypothesis posits that positive interactions increase and competitive interactions decrease with increasing consumer pressure or environmental stress. This hypothesis has received extensive attention in plant community ecology, but only a handful of tests in animals. Furthermore, few empirical studies have examined multiple co‐occurring stressors. Here we test predictions of the stress gradient hypothesis using the occurrence of mixed‐species groups in six common grazing ungulate species within the Serengeti‐Mara ecosystem. We use mixed‐species groups as a proxy for potential positive interactions because they may enhance protection from predators or increase access to high‐quality forage. Alternatively, competition for resources may limit the formation of mixed‐species groups. Using more than 115,000 camera trap observations collected over 5 yr, we found that mixed‐species groups were more likely to occur in risky areas (i.e., areas closer to lion vantage points and in woodland habitat where lions hunt preferentially) and during time periods when resource levels were high. These results are consistent with the interpretation that stress from high predation risk may contribute to the formation of mixed‐species groups, but that competition for resources may prevent their formation when food availability is low. Our results are consistent with support for the stress gradient hypothesis in animals along a consumer pressure gradient while identifying the potential influence of a co‐occurring stressor, thus providing a link between research in plant community ecology on the stress gradient hypothesis, and research in animal ecology on trade‐offs between foraging and risk in landscapes of fear.

Refining the stress gradient hypothesis for mixed species groups of African mammals

“…Thus, if aggregating with other species has fitness relevant benefits, the SGH predicts that prey species associate with other species in areas where they are subject to greater predation risk. While this prediction has been supported in quantitative terms, exemplified by MSGs of grazing herbivores more likely to occur in areas with high lion ( Panthera leo ) predation risk 9 , we additionally predict that species associations change in composition. Such structural changes could arise in areas with high predation risk, if species associate with heterospecifics with complementary modes of predator detection to increase predator detection or with a similar susceptibility to predation to dilute individual risk.…”

“…If animal associations followed predictions of the SGH, we would expect that MSGs form most frequently under conditions of resource scarcity. However, quantitative tests of the SGH within animal communities did not always support this prediction 13 and in East African herbivore communities, species associations were often positively associated with primary productivity 7 , 9 . While these patterns may speciously contradict predictions of the SGH, the directionality of stress associated with primary productivity may likely differ across species and could be mediated by body size and digestive strategy 23 , but Ref.…”

“…Individuals from different species may associate for several reasons, and variation in predation risk and resource availability have been hypothesized to mediate the frequency and structure of MSGs 4 – 8 . More recently, the formation of MSGs has been linked to the stress gradient hypothesis (SGH) 9 . The SGH postulates an increase in positive species interactions with increasing environmental stress 10 , 11 .…”

A case for associational resistance: Apparent support for the stress gradient hypothesis varies with study system

“…The SGH has been investigated and debated by plant community ecologists since its conception, but it is not as well known outside of plant community ecology (Barrio et al, 2013; Beaudrot et al, 2020; Daleo & Iribarne, 2009; Dangles et al, 2013). In other ecological fields, studies of the effects of stress may indicate support for or against the SGH without acknowledging the hypothesis’ existence, and it is unclear how much the nature of the ecosystem and the stress itself affects conclusions (Lortie & Callaway, 2006; Maestre et al, 2006).…”

Bird species co‐occurrence patterns in an alpine environment supports the stress‐gradient hypothesis

“…In this sense, the stress‐gradient hypothesis (SGH) predicts that the frequency of positive and negative interactions will vary inversely across abiotic stress gradients, with species aggregations being more common in conditions of high abiotic stress (less productive systems) relative to more benign conditions (Bertness and Callaway 1994, Götzenberger et al 2012). Although it has been proposed that this theoretical framework, originally formulated by plant ecologists, might also be extended to higher trophic levels including terrestrial vertebrates, few studies so far have tested the stress (productivity)‐gradient hypothesis in animal communities (Barrio et al 2013, Beaudrot et al 2020).…”