Toggle Main Menu Toggle Search

Open Access padlockePrints

Ternary network models for disturbed ecosystems

Lookup NU author(s): Dr Kieran Peel, Professor Darren Evans, Dr Clive Emary



This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).


The complex network of interactions between species makesunderstanding the response of ecosystems to disturbances anenduring challenge. One commonplace way to deal with thiscomplexity is to reduce the description of a species to a binarypresence‚Äďabsence variable. Though convenient, this limits thepatterns of behaviours representable within such models. Weaddress these shortcomings by considering discrete populationmodels that expand species descriptions beyond the binarysetting. Specifically, we focus on ternary (three-state) modelswhich, alongside presence and absence, additionally permitspecies to become overabundant. We apply this ternaryframework to the robustness analysis of model ecosystems andshow that this expanded description permits the modelling oftop-down extinction cascades emerging from consumerpressure or mesopredator release. Results therefore differsignificantly from those seen in binary models, where sucheffects are absent. We also illustrate how this method opens upthe modelling of ecosystem disturbances outside the scope ofbinary models, namely those in which species are externallyraised to overabundance. Our method therefore has thepotential to provide a richer description of ecosystem dynamicsand their disturbances, while at the same time preserving theconceptual simplicity of familiar binary approaches.

Publication metadata

Author(s): Peel K, Evans DM, Emary C

Publication type: Article

Publication status: Published

Journal: Royal Society Open Science

Year: 2022

Volume: 9

Issue: 10

Print publication date: 26/10/2022

Online publication date: 26/10/2022

Acceptance date: 30/09/2022

Date deposited: 31/10/2022

ISSN (electronic): 2054-5703

Publisher: The Royal Society Publishing


DOI: 10.1098/rsos.220619


Altmetrics provided by Altmetric


Funder referenceFunder name