Skip to main content
Dryad logo

The effects of predation-risk on prey stoichiometry: A meta-analysis

Citation

Rinehart, Shelby; Hawlena, Dror (2021), The effects of predation-risk on prey stoichiometry: A meta-analysis, Dryad, Dataset, https://doi.org/10.25338/B8D909

Abstract

A decade ago, the General Stress Paradigm (GSP) aimed to develop a predictive framework linking predator effects to ecosystem function. The GSP was based on the notion that animals, across taxa, exhibit similar physiological responses to predation risk that divert resources from growth and reproduction [which require N-rich biomolecules (i.e., protein)] to emergency functions [which require C-rich biomolecules (e.g., carbohydrates)]. The GSP predicts that stressed prey should have a greater dietary demand for C-rich resources, a higher body C:N ratio, and elevated N excretion. Now, ten years later, we aim to revisit the GSP— using quantitative meta-analysis to test the original predictions of the GSP and how 1) predator hunting mode, 2) multiple stressors, and 3) prey dietary shifts affect prey stoichiometric responses to predation-risk. Our dataset was consistent with previous work showing that predation-risk increases prey glucocorticoid levels and metabolic rates and decreases prey growth rates. We found that predation-risk tended to decrease the fat, carbohydrate, and protein content of prey bodies; increased the C:P and N:P of prey bodies; but had no effect on the C, N, P or C:N content of prey bodies. Additionally, we found no effect of predation risk on the N content of prey excretions. Prey responses to predation risk were unaffected by multiple stressors or the prey’s ability to shift their diet, but predator hunting mode did affect the nature of prey stoichiometric responses. Specifically, ambush predators decreased prey macronutrient content and suppressed prey growth; while active predators had no effect on prey macronutrient content and a smaller effect on prey growth than ambush predators.  The significant effects of predation risk were supported by robust fail-safe numbers, despite the high between comparison heterogeneity that was found in all analyses. Our findings highlight the need to expand the GSP framework to include more diverse taxa and ecosystems as well as other types of inducible defenses; while also considering predator traits and the mismatch between macronutrient content and prey stoichiometry.

Methods

We surveyed the literature using Web of Science and the following search terms, "predat* AND prey stoichiometry OR prey physiological stress OR prey nutritional geometry OR prey general stress paradigm OR prey physiological plasticity." The search was conducted on 10 June 2019. We used the preferred reporting practices outlined by PRISMA to structure our overall literature search (Moher et al. 2009). Our search identified 435 potential manuscripts. For each potential manuscript, we read the abstract and determined if the study measured prey stoichiometric responses to predation-risk (including conspecific alarm cues). This screening yielded 51 manuscripts that we read in full to determine if they were eligible for inclusion in our quantitative meta-analysis (see Fig. 1). Studies were deemed eligible if they induced one or more measurement of prey stoichiometry (i.e., macronutrient or elemental composition) in the presence and absence of predation-risk. We targeted studies that measured the effects of predation-risk on: (i) the macronutrient composition, (ii) the elemental composition, and (iii) the elemental ratios of a prey’s body and waste materials. After reading 51 manuscripts in full, we had to exclude 24 manuscripts because they either failed to address our objectives or were missing data (e.g., sample size, Fig. 1). From the 27 included manuscripts, we also extracted data on the effects of predation-risk on prey growth, foraging, assimilation efficiency, metabolic rate, RNA: DNA, and CORT concentrations when possible. We extracted these additional metrics because previous meta-analyses on predator effects have generated robust predictions for how predation-risk should affect these prey traits (see Preisser et al. 2005, Preisser and Bolnick 2008). Thus, by comparing the findings from our dataset to these pre-existing conclusions (based on larger datasets) we hoped to gain insight into how studies in our small dataset compare to the larger body of literature on predator-risk effects.

Usage Notes

NA in predator hunting mode means study used conspecific cuses rather than direct predator cues. Unknown in predator hunting mode means we were unable to classify the predator's hunting mode based in the information provided in the study or from outside resources. 

Funding

Minerva Center for Movement Ecology, Hebrew University of Jerusalem, Award: Postdoctoral Research Fellowship to SR

Zuckerman STEM Leadership Fellowship, Award: Postdoctoral Research Fellowship to SR

Lady Davis Fellowship Trust, Hebrew University of Jerusalem, Award: Postdoctoral Research Award to SR

FP7 Ideas: European Research Council, Award: Award to DH