Experimental reduction of plant abundance changes interaction frequency of a tri‐trophic micro‐food web: contrasting responses of generalists and specialists

Abstract

Species abundance is vulnerable to climate change and anthropogenic impact. Although numerous studies have examined the food web response to species loss, their response (e.g. in network topology and interaction frequency) to changes in species abundance has received little attention. We experimentally reduced the abundance (by c. 60%) of one host plant species (Saussurea nigrescens) in a tri‐trophic micro‐food web consisting of two common Asteraceae species (S. nigrescens and Anaphalis flavescens), their pre‐dispersal seed predators (tephritid flies) and the parasitoid wasps that feed on these seed predators. The neutral process hypothesis posits that network topology and interaction frequency are determined by the relative abundances of the participating species. Accordingly, we hypothesized that interaction frequency (indicated by the rate of flies infesting plants and of parasitoid wasps parasitizing flies) would decrease with decreasing abundance of host plant abundance. Consistent with the neutral process hypothesis, abundance reduction significantly decreased the rate of generalist flies infesting the manipulated plant host species, but increased the rate on unmanipulated plant species. In contrast, abundance reduction significantly increased the rate of specialist flies infesting the manipulated species, but it had no demonstrable effect on the specialists on the unmanipulated plant host. Moreover, abundance reduction significantly increased the rate of parasitoid wasps (as a species group) parasitizing flies in the manipulated species, but decreased the rate of the unmanipulated species. These results were not attributable to neutral processes, but can be explained by adaptive foraging theory. In addition, experimental manipulation did not shift the qualitative presence–absence of the linkages in the micro‐food web, and the change in the abundance of both fly and wasp species was smaller than the abundance changes of the manipulated plant species. Synthesis. Our results indicate that both neutral processes and/or adaptive foraging likely contributed to food web stability in responses to changes in species abundance. Because species abundance is vulnerable to abiotic and biotic environmental changes, we suggest more research should be conducted to understand how food webs respond to climate change and anthropogenic disturbance.

Plant volatiles mediate evolutionary interactions between plants and tephritid flies and are evolutionarily more labile than non‐volatile defenses

“…Host associations between plants and tephritid flies were determined in 2017 by collecting mature capitula and identifying emerged adults by their morphology, or by DNA barcoding larval flies within capitula (Xi et al., 2019). A total of 73 interactions involving 17 plant species and 20 associated fly species were identified in this study (Figure S1).…”

“…A total of 73 interactions involving 17 plant species and 20 associated fly species were identified in this study (Figure S1). This interaction network was a part of a larger network that included 24 plant species and 23 fly species (Xi et al., 2019).…”

Galling insect communities mediate the effects of fire on their associated parasitoid communities

“…Food webs represent how several species interact (Poisot et al, 2016) and provide information about the structure and functioning of communities, which can be overlooked when they are studied only with diversity metrics (Lewinsohn et al, 2022; Tylianakis et al, 2007). For example, variations in the host–parasitoid food web structure can be observed when parasitoid species behave as specialists or generalists, depending on the abundance of their preferred hosts (Barbosa et al, 2017; Ferracini et al, 2018; Schönrogge & Crawley, 2000; Tylianakis et al, 2007; Xi et al, 2020). Thus, assuming that abundance and species composition of galling insects are affected by fire, a change in the network structure is expected in different fire scenarios (Bakhshandeh‐Savadroodbari et al, 2017; Cronin et al, 2020; Kuzmanich et al, 2022).…”