Quantifying interaction networks and stability properties of plankton food webs using multivariate first order autoregressive modelling

A.S. Gsell, Deniz Özkundakci, Marie-Pier Hébert, Rita Adrian

Research output: Chapter in book/volumeChapterScientific

22 Downloads (Pure)

Abstract

Lakes and reservoirs have been identified as sentinels of global change as they integrate
changes in the surrounding landscape. While univariate indicator variables are relatively well assessed,
the lack of knowledge on temporal changes in species interactions under pressure has been identified as
a major gap in the bio-monitoring sciences. Multivariate autoregressive models can be used to assess
direction and strength of both direct and indirect interactions in complex communities over time. This
model framework also allows calculation of network stability properties (variance, resilience and reactivity). Moreover, the interaction matrix can be further analyzed for classical network structure properties (closeness- and betweenness centrality). These measures are useful indicators of changes in ecosystem stability and help identify biotic keystone groups and/or groups of species that are particularly
vulnerable to changes in the landscape
Original languageEnglish
Title of host publicationNovel Methods and Results of Landscape Research in Europe, Central Asia and Siberia
Subtitle of host publicationVol. 3. Landscape Monitoring and Modelling
EditorsViktor Sychev, Lothar Mueller
Place of PublicationMoscow
PublisherRussian Academy of Sciences
Pages306-309
Volume3
ISBN (Print)978-5-9238-0249-8
DOIs
Publication statusPublished - 2018

    Fingerprint

Keywords

  • international

Cite this

Gsell, A. S., Özkundakci, D., Hébert, M-P., & Adrian, R. (2018). Quantifying interaction networks and stability properties of plankton food webs using multivariate first order autoregressive modelling. In V. Sychev, & L. Mueller (Eds.), Novel Methods and Results of Landscape Research in Europe, Central Asia and Siberia : Vol. 3. Landscape Monitoring and Modelling (Vol. 3, pp. 306-309). Russian Academy of Sciences. https://doi.org/10.25680/3366.2018.41.77.257