Review
Inferring biotic interactions from proxies

https://doi.org/10.1016/j.tree.2015.03.014Get rights and content

Highlights

  • We review some of the challenges associated with predicting biotic interactions.

  • We propose a framework for predicting interactions between species based on functional traits, phylogenies, and geography.

  • We illustrate the predictive power of the framework with published food-web data.

  • We conclude that there is scope to see the ‘forest beyond the trees’ and make sense of complexity in the web of life.

Inferring biotic interactions from functional, phylogenetic and geographical proxies remains one great challenge in ecology. We propose a conceptual framework to infer the backbone of biotic interaction networks within regional species pools. First, interacting groups are identified to order links and remove forbidden interactions between species. Second, additional links are removed by examination of the geographical context in which species co-occur. Third, hypotheses are proposed to establish interaction probabilities between species. We illustrate the framework using published food-webs in terrestrial and marine systems. We conclude that preliminary descriptions of the web of life can be made by careful integration of data with theory.

Section snippets

Why infer interactions?

Even if serious gaps in knowledge of biodiversity remain, much progress has been made in determining how many different types of organisms exist (the Linnaean shortfall [1]), what evolutionary relationships connect different lineages to a common ancestor (the Darwinian shortfall [2]), and where different species are distributed (the Wallacean shortfall [3]). Much less is known about the types of interactions that exist among species (the Eltonian shortfall [4]) and the importance of such

Which interactions to infer?

There are many different ways to describe a biotic interaction. Interactions may vary in their type (e.g., antagonistic or facilitative), their strength (e.g., weak or strong interactions), or their symmetry (e.g., symmetric or asymmetric). An important step for inferring biotic interactions is to determine what information is to be inferred. We propose building interaction networks bottom-up, in other words predicting the links among species and then exploring the collective properties of the

The proxies used in inference

When direct information about biotic interactions is unavailable, we must resort to indirect information or proxies to obtain insight about them. Three classes of proxies can help with inferring interactions between species: traits, phylogenies, and geographical data (for a review of examples see Table 1). Traits are usually defined as morphological, physiological, phenological, or behavioral characteristics of species that directly impact on their fitness [25]. However, they are also expected

Testing inferences about interactions

The usefulness of a theoretical model is partly dependent on it being successfully tested. However, inferences of biotic interactions by models are not easily tested because reliable data on absence of interactions are generally unavailable. Similar problems exist in the literature on modeling species distributions 4, 55, with the consequence that inferences of interactions must necessarily be interpreted as potential rather than realized. Indeed, observed interactions will typically constitute

Concluding remarks

We have proposed a framework for inferring biotic interactions based on stepwise removal of forbidden links and calculation of the probabilities of interaction for the remaining links. With such a process one is able to establish the backbone of an interactions network occurring in a given regional species pool. The pruning of the network is made using rules derived from the analysis of functional traits, phylogenies, and geographical proxies. To provide an illustration we implemented the

Acknowledgments

We thank Natalia Melo for support with the design of the artwork. Research by I.M-C., M.G.M., and M.B.A. was supported through the Integrated Program of Investigação Cientifica e Desenvolvimento Tecnológico (IC&DT) (1/SAESCTN/ALENT-07-0224-FEDER-001755). I.M-C., is currently funded by the Fonds de Recherches du Québec – Nature et Technologies (FQRNT) programme. M.G.M. also acknowledges support by a Marie Curie Intra-European Fellowship within the 7th European Commission Framework Programme

References (81)

  • M. Loreau

    From Populations to Ecosystems: Theoretical Foundations for a New Ecological Synthesis

    (2010)
  • M.B. Araújo et al.

    The importance of biotic interactions for modelling species distributions under climate change

    Global Ecol. Biogeogr.

    (2007)
  • R.K. Heikkinen

    Biotic interactions improve prediction of boreal bird distributions at macro-scales

    Global Ecol. Biogeogr.

    (2007)
  • B.L. Bateman

    Biotic interactions influence the projected distribution of a specialist mammal under climate change. Diversity and Distributions

    (2012)
  • N.J. Gotelli

    Macroecological signals of species interactions in the Danish avifauna

    Proceedings of the National Academy of Sciences

    (2010)
  • M.B. Araújo

    Using species co-occurrence networks to assess the impacts of climate change

    Ecography

    (2011)
  • J.M. Montoya et al.

    Climate change, biotic interactions and ecosystem services

    Philosophical Transactions of the Royal Society B: Biological Sciences

    (2010)
  • J.L. Blois

    Space can substitute for time in predicting climate-change effects on biodiversity

    Proc. Natl. Acad. Sci. U.S.A.

    (2013)
  • J. Bascompte

    Disentangling the web of life

    Science

    (2009)
  • N.D. Martinez

    Artifacts or attributes? Effects of resolution on the Little Rock Lake food web

    Ecol. Monogr.

    (1991)
  • T. Poisot

    Beyond species: why ecological interactions vary through space and time

    Oikos

    (2014)
  • M.B. Araújo et al.

    The geographic scaling of biotic interactions

    Ecography

    (2014)
  • M.S. Laska et al.

    Theoretical concepts and empirical approaches to measuring interaction strength

    Ecology

    (1998)
  • J. Bascompte

    Asymmetric coevolutionary networks facilitate biodiversity maintenance

    Science

    (2006)
  • J.T. Wootton

    Estimates and tests of per capita interaction strength: diet, abundance, and impact of intertidally foraging birds

    Ecol. Monogr.

    (1997)
  • P.C.d. Ruiter

    Calculation of nitrogen mineralization in soil food webs

    Plant Soil

    (1993)
  • C. Violle

    Let the concept of trait be functional!

    Oikos

    (2007)
  • N. Mouquet

    Ecophylogenetics: advances and perspectives

    Biol. Rev.

    (2012)
  • S. Kéfi

    More than a meal… integrating non-feeding interactions into food webs

    Ecol. Lett.

    (2011)
  • M.E.J. Newman

    Networks: An Introduction

    (2010)
  • D. Simberloff et al.

    The guild concept and the structure of ecological communities

    Annu. Rev. Ecol. Syst.

    (1991)
  • E.B. Baskerville

    Spatial guilds in the Serengeti food web revealed by a Bayesian group model

    PLoS Comp. Biol.

    (2011)
  • P.D. Roopnarine et al.

    Detailed food web networks of three Greater Antillean coral reef systems: the Cayman Islands, Cuba, and Jamaica

    Dataset Papers Ecol.

    (2013)
  • A. Eklöf

    The dimensionality of ecological networks

    Ecol. Lett.

    (2013)
  • M.B. Bonsall et al.

    Apparent competition structures ecological assemblages

    Nature

    (1997)
  • R.D. Holt et al.

    Apparent competition and enemy-free space of insect host-parasitoid communities

    Am. Nat.

    (1993)
  • E. Borer

    What determines the strength of a trophic cascade?

    Ecology

    (2005)
  • D.A. Vasseur et al.

    A mechanistic approach for modeling temperature-dependent consumer resource dynamics

    Am. Nat.

    (2005)
  • T. Poisot

    The dissimilarity of species interaction networks

    Ecol. Lett.

    (2012)
  • B. Lavandero et al.

    Genotype matching in a parasitoid–host genotypic food web: an approach for measuring effects of environmental change

    Mol. Ecol.

    (2013)
  • Cited by (243)

    View all citing articles on Scopus
    View full text