Progressive sensitivity of trophic levels to warming underlies an elevational gradient in ant–aphid mutualism strength

Richard Alexander Smith,Annika S Nelson,Kailen A Mooney,Riley T Pratt,Cole T Symanski,Cathrine Prenot,Jessica D Pratt

doi:10.1111/oik.05650

Abstract

Although species interactions are often proposed to be stronger at lower latitudes and elevations, few studies have evaluated the mechanisms driving such patterns. In this study, we assessed whether, and by which mechanisms, abiotic changes associated with elevation altered the outcome of an ant–aphid protection mutualism. To do so, we characterized the multi‐trophic interactions among the ant Formica podzolica, the aphid Aphis varians, and aphid natural enemies occurring on the plant Chamerion angustifolium within replicate high and low elevation valleys. Low (versus high) elevation sites had longer summers (snowmelt 13 days earlier) and were on average 1.1°C warmer and 41% drier throughout the year. At low elevations, individual ant colonies consumed approximately double the volume of carbohydrate baits, likely due to a higher foraging tempo, and possibly due to a greater demand for sugar‐ versus protein‐rich resources (as indicated by stable isotope analysis). Wild aphid colonies at low elevations were visited by 1.4‐fold more natural enemies (controlling for variation in aphid abundance), while experimental aphid colonies on potted plants were tended 52% more frequently by ants. As a result, ants increased aphid colony survival by 66% at low elevations but had no detectable effect at high elevations; at low (versus high) elevations aphid colonies without ants had lower survival, demonstrating stronger predator effects, while aphid colonies with ants had higher survival, demonstrating even stronger ant benefits. Analyses for the effects of mean summer temperature yielded qualitatively identical results to those based on elevation. Collectively, these findings support predictions for a greater sensitivity of higher trophic levels to warming and demonstrate how species interactions can vary across environmental gradients due to simultaneous changes in species traits and abundances across multiple trophic levels.

Highlights

Species interactions, from antagonisms to mutualisms, have been proposed to be stronger at lower elevations and towards the tropics, where temperatures are warmer and the climate is less seasonal (Schemske et al 2009, Moreira et al 2018)
These results are consistent with predictions for stronger species interactions at lower elevations and for the progressive sensitivity of trophic levels to temperature
Low elevation sites were associated with a higher natural enemy abundance, and in the absence of ants, lower aphid colony survival, supporting the prediction for stronger predator effects under warmer conditions

Summary

Introduction

From antagonisms to mutualisms, have been proposed to be stronger at lower elevations and towards the tropics, where temperatures are warmer and the climate is less seasonal (Schemske et al 2009, Moreira et al 2018). Recent studies have questioned the generality of this pattern (Moles et al 2011, Moles and Ollerton 2016), and we lack a clear mechanistic framework for predicting variation in species interactions along environmental gradients (Moreira et al 2018). Understanding the factors that cause variation in species interactions is increasingly important for predicting how communities will be affected by climate change (Tylianakis et al 2008, Gilman et al 2010). Pairwise interactions should vary if the abiotic environment directly affects the traits or abundances of one or both interacting species. Gradients in pairwise interactions may be driven by environmental effects on the focal species, and by variation in the multi-trophic food webs within which the interactions are embedded (Walther 2010, Mooney et al 2016). A mechanistic understanding of gradients in species interactions requires documenting variation in interaction outcomes, and in the abundances and traits of the interacting species, and the biotic contexts within which their interactions occurs (Gilman et al 2010, Mooney et al 2016)

Methods

Results

Conclusion