Sources of prey availability data alter interpretation of outputs from prey choice null networks Number of Authors: 82024 (English)In: Ecological Entomology, ISSN 0307-6946, E-ISSN 1365-2311, Vol. 49, no 3, p. 418-432Article in journal (Refereed) Published

Abstract [en]

1. Null models provide a valuable baseline against which fundamental ecological hypotheses can be tested and foraging choices that cannot be explained by neutral processes or sampling biases can be highlighted. In this way, null models can advance our understanding beyond simplistic dietary descriptions to identify drivers of interactions. This method, however, requires estimates of resource availability, which are generally imperfect representations of highly dynamic systems. Optimising method selection is crucial for study design, but the precise effects of different resource availability data on the efficacy of null models are poorly understood.
2. Using spider–prey networks as a model, we used prey abundance (suction sample) and activity density (sticky trap) data, and combinations of the two, to simulate null networks. We compared null diet composition, network properties (e.g., connectance and nestedness) and deviations of simulations from metabarcoding-based spider dietary data to ascertain how different prey availability data alter ecological interpretation.
3. Different sampling methods produced different null networks and inferred distinct prey selectivity. Null models based on prey abundance and combined frequency-of-occurrence data generated null diet compositions, which more closely resembled the diet composition determined by metabarcoding. Null models based on prey abundance, activity density and proportionally combined data generated null network properties most like the networks constructed via dietary metabarcoding.
4. We show that survey method choice impacts all aspects of null network analyses, the precise effects varying between methods but ultimately altering ecological interpretation by increasing disparity in network properties or trophic niches between null and directly constructed networks. Merging datasets can generate more complete prey availability data but is not a panacea because it introduces different biases. The choice of method should reflect the research hypotheses and study system being investigated. Ultimately, survey methods should emulate the foraging mode of the focal predator as closely as possible, informed by the known ecology, natural history and behaviour of the predator.

Place, publisher, year, edition, pages

2024. Vol. 49, no 3, p. 418-432

Keywords [en]

dietary analysis, high-throughput sequencing, metabarcoding, network ecology, null modelling, resource preference

National Category

Ecology

Identifiers

URN: urn:nbn:se:su:diva-226644DOI: 10.1111/een.13315ISI: 001154970000001Scopus ID: 2-s2.0-85184686934OAI: oai:DiVA.org:su-226644DiVA, id: diva2:1837964

Available from: 2024-02-15 Created: 2024-02-15 Last updated: 2024-11-13Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Hambäck, Peter A.

Search in DiVA

By author/editor

Hambäck, Peter A.

By organisation

Department of Ecology, Environment and Plant Sciences

In the same journal

Ecological Entomology

On the subject

Ecology

Search outside of DiVA

GoogleGoogle ScholarTotal: 58 hits