Elsevier

Theoretical Population Biology

Volume 89, November 2013, Pages 24-33
Theoretical Population Biology

Coexistence of structured populations with size-based prey selection

https://doi.org/10.1016/j.tpb.2013.07.003Get rights and content

Abstract

Species with a large adult–offspring size ratio and a preferred predator–prey mass ratio undergo ontogenetic trophic niche shift(s) throughout life. Trophic interactions between such species vary throughout life, resulting in different species-level interaction motifs depending on the maximum adult sizes and population size distributions. We explore the assembly and potential for coexistence of small communities where all species experience ontogenetic trophic niche shifts. The life-history of each species is described by a physiologically structured model and species identity is characterised by the trait: size at maturation. We show that a single species can exist in two different states: a ‘resource driven state’ and a ‘cannibalistic state’ with a large scope for emergent Allee effects and bistable states. Two species can coexist in two different configurations: in a ‘competitive coexistence’ state when the ratio between sizes at maturation of the two species is less than a predator–prey mass ratio and the resource level is low to intermediate, or in a ‘trophic ladder’ state if the ratio of sizes at maturation is larger than the predator–prey mass ratio at all resource levels. While there is a large scope for coexistence of two species, the scope for coexistence of three species is limited and we conclude that further trait differentiation is required for coexistence of more species-rich size-structured communities.

Introduction

The ecological role of fish in aquatic communities is shaped by two characteristics: (1) individuals have a size-based preference for prey 10–1000 times smaller than themselves in terms of weight (Brose et al., 2006, Barnes et al., 2010), and (2) they have a life-history where they produce many small offspring, around 1 mg, which is several orders of magnitude smaller than their adult size ranging from 1 g to 100 kg (Duarte and Alcaraz, 1989, Winemiller and Rose, 1993, Andersen et al., 2008). Consequently this means that individuals feed on prey from successively higher trophic levels throughout ontogeny: as larvae individuals compete for shared resources of zooplankton nauplii, and as they grow they shift to adult zooplankton and later to piscivory and feeding on larger invertebrates. Individuals therefore undergo ontogenetic niche shifts (Werner and Gilliam, 1984) and specifically shifts in the trophic niche (also known as life-history omnivory; Pimm and Rice, 1987). Due to ontogenetic trophic niche shifts two individuals may interact either competitively, if they are of a similar size, or as predator and prey if the size of one individual falls into the preferred prey size range of the other. The ecological role is therefore determined by the size of an individual rather than by its species identity. This complication makes it difficult to characterise the interaction between two fish populations as “competitive” or “predator–prey” since the species-level interaction type is likely to be a complex mixture of both that depends on the size structure of the two populations (Fig. 1).

The classic way of describing species-level interactions is through unstructured models based on coupled ordinary differential equations (e.g. MacArthur and Levins, 1967; Oksanen et al., 1981; Holt and Polis, 1997). In unstructured models energy is transferred between species only through predator–prey interactions. In populations with ontogenetic trophic niche shifts energy is also transferred through somatic growth in body size. Explicit modelling of ontogenetic trophic niche shifts therefore requires the use of physiologically structured modelling approaches (Metz and Diekmann, 1986, de Roos and Persson, 2001, de Roos and Persson, 2013). The key characteristic of physiologically structured models is that they resolve food-dependent growth and reproduction, where growth can either be between stages (de Roos et al., 2008a), in a cohort model with several state variables (de Roos and Persson, 2001), or with a continuous size structure (Claessen and de Roos, 2003, Hartvig et al., 2011). The dependence of growth and reproduction on food encounter is what distinguishes physiologically structured models from the traditional age-based models used in fisheries science (Beverton and Holt, 1957). Analysis of physiologically structured models have e.g. demonstrated coexistence of several unstructured predator populations on a single structured prey population through the mechanism of emergent facilitation (de Roos et al., 2008b). It seems, however, that inclusion of ontogenetic trophic niche shifts decreases the possibility of coexistence of two size-structured species (van de Wolfshaar et al., 2006).

We explore the potential for coexistence of species experiencing ontogenetic trophic niche shifts. To simplify matters we assume that selection of food is determined solely by the differences in body size of the interacting individuals—meaning that we disregard any dietary selection based on species identity. We refer to such populations as “purely size-structured”. We apply a continuously size-structured approach where individual physiological processes, including predator–prey encounter, are parameterised by relations with the size of individuals. In this manner each individual is characterised by body size, while each species is characterised by a single trait: the size at maturation (Hartvig et al., 2011). The formulation of purely size-structured populations through a single trait allows for a general analysis of coexistence as variation of a larger set of species-specific parameters are avoided. The analysis is divided into three steps: we first characterise a resource–consumer system where one structured species feeds on a size-structured resource and potentially on itself through cannibalism. In this system we identify the resource-driven and cannibalistic population states described earlier (Claessen and de Roos, 2003) including an emergent Allee effect (de Roos et al., 2003) and bistable states (Claessen and de Roos, 2003, van Kooten et al., 2005, Guill, 2009). Next we show how two species may coexist. Depending on the relative sizes at maturation the two species can be considered part of different species-level interaction motifs. If the sizes at maturation are similar, the two species will mainly interact as competitors for the resource. If size at maturation of one species is significantly larger than the other species the interaction is a combination of competitive coexistence and predator–prey interactions (including cannibalism). Finally we explore the potential for coexistence of three or more species and show how the scope for coexistence decreases rapidly when more species are introduced into the system.

Section snippets

Model

We employ a recent size- and trait-based model framework to describe the population dynamics of purely size-structured interacting populations (Hartvig et al., 2011). The model differs from the Hartvig et al. (2011) model only by assuming that interaction strengths between species is solely due to size-dependent predation (i.e. setting the species-specific couplings strengths to unity). The model formalises the sketch in Fig. 1 by explicitly resolving the entire individual life-history from

Methods

The model (1) consists of coupled partial integro-differential equations and is solved numerically using a first order semi-implicit upwind finite-difference scheme (Press et al., 1992, Hartvig et al., 2011). The size-spectrum of all species is discretised on a mass grid m0m85kg with 200 logarithmically even sized mass (m) groups, and the time step used for integration is 0.02 years. The dynamics of the resource spectrum may be obtained analytically at each time step to save computational

Results

The analysis of a system with just one species demonstrates (Fig. 2): (1) There is an upper boundary to the size at maturation m of a single species that can invade from small numbers (thick solid line). This maximum size of m is determined by the resource level described by the initial feeding level f0; higher resource levels leads to larger maximum size. (2) Beyond the invasion boundary species may persist but they cannot invade; these states can be reached by starting simulations from a

Discussion

Single species states

A single species with ontogenetic trophic niche shifts living on a resource is known to have the potential for several stable states (Claessen and de Roos, 2003, de Roos et al., 2003, van Kooten et al., 2005). These states are also found here and can be understood in terms of an energetic bottleneck. Through invasion in small numbers a resource-driven state may be reached where the population primarily feed on the resource spectrum. The maximum body size a species can

Acknowledgements

M.H. acknowledges support from European Marie Curie RTN FishACE through the EU’s 6th FP (MRTN-CT-2004-005578) and the Danish National Research Foundation for support to the Center for Macroecology, Evolution and Climate. K.H.A. was supported by the Villum Kann Rasmussen Centre of Excellence: Ocean Life, and the EU’s 7th framework research projects FACTS and MYFISH.

References (63)

  • P. Amarasekare

    Coexistence of intraguild predators and prey in resource-rich environments

    Ecology

    (2008)
  • K.H. Andersen et al.

    Asymptotic size determines species abundance in the marine size spectrum

    Am. Nat.

    (2006)
  • M. Arim et al.

    Intraguild predation: a widespread interaction related to species biology

    Ecol. Lett.

    (2004)
  • C. Barnes et al.

    Global patterns in predator–prey size relationships reveal size dependency of trophic transfer efficiency

    Ecology

    (2010)
  • R.J.H. Beverton et al.
  • U. Brose et al.

    Consumer-resource body-size relationships in natural food webs

    Ecology

    (2006)
  • P. Chesson et al.

    The interaction between predation and competition

    Nature

    (2008)
  • D. Claessen et al.

    The impact of size-dependent predation on population dynamics and individual life history

    Ecology

    (2002)
  • J.E. Cohen et al.

    Body sizes of animal predators and animal prey in food webs

    J. Anim. Ecol.

    (1993)
  • A.M. de Roos et al.

    Physiologically structured models—from versatile technique to ecological theory

    Oikos

    (2001)
  • A.M. de Roos et al.

    Population and Community Ecology of Ontogenetic Development, Vol. 51

    (2013)
  • A.M. de Roos et al.

    Emergent Allee effects in top predators feeding on structured prey populations

    Proc. R. Soc. Lond. B

    (2003)
  • A.M. de Roos et al.

    Stage-specific predator species help each other to persist while competing for a single prey

    Proc. Natl. Acad. Sci. USA

    (2008)
  • S. Diehl et al.

    Effects of enrichment on three-level food chains with omnivory

    Am. Nat.

    (2000)
  • C.M. Duarte et al.

    To produce many small or few large eggs: a size-independent reproductive tactic of fish

    Oecologia

    (1989)
  • M. Elgar et al.

    Cannibalism: Ecology and Evolution Among Diverse Taxa

    (1992)
  • T. Fenchel

    Intrinsic rate of natural increase: the relationship with body size

    Oecologia

    (1974)
  • L. Fox

    Cannibalism in natural populations

    Annu. Rev. Ecol. Evol. Syst.

    (1975)
  • S.A.H. Geritz et al.

    Invasion dynamics and attractor inheritance

    J. Math. Biol.

    (2002)
  • Hartvig, M., 2011. Ecological processes yield complex and realistic food webs. In: Food Web Ecology—Individual...
  • V. Hin et al.

    Coexistence of predator and prey in intraguild predation systems with ontogenetic niche shifts

    Am. Nat.

    (2011)
  • Cited by (32)

    • A general size- and trait-based model of plankton communities

      2020, Progress in Oceanography
      Citation Excerpt :

      The general result of IGP is that the consumer outcompetes the predator at low levels of productivity. At medium productivities the predator and consumer coexist, while the predator dominates at high productivities (Mylius et al., 2001; Hartvig and Andersen, 2013). This result is hard to track in the model since IGP occurs at several levels simultaneously, e.g., between protists and small copepods, and between small copepods and larger copepods.

    • Trait-based modelling in ecology: A review of two decades of research

      2019, Ecological Modelling
      Citation Excerpt :

      An adaptation and a dynamic version of the model of a theoretical fish community (Pope et al., 2006), based on classical multi-species fishery models and community size spectrum models, was reconsidered by Andersen and Pedersen (2010) and Andersen et al. (2015) in the framework of a trait-based approach. A similar model to describe population structure based on the size of the organisms was presented by Hartvig et al. (2011) and Hartvig and Andersen (2013). This model was a product of the synthesis between traditional unstructured food webs, allometric body size scaling, trait-based modelling, and physiologically structured modelling (Hartvig et al., 2011).

    • From Bacteria to Whales: Using Functional Size Spectra to Model Marine Ecosystems

      2017, Trends in Ecology and Evolution
      Citation Excerpt :

      This advance makes SSMs similar to physiologically structured population models but capable of resolving the complexity of food web models through the use of traits [4,50,51,53]. Increasing trait diversity in the SSMs has the effect of stabilizing complex ecological communities [54–56]. Multispecies extensions of the trait-based model have enabled real-world ecosystem applications in combination with extensive species-specific trait data for parameterization [57–60] (Box 1).

    • Modelling the community size-spectrum: recent developments and new directions

      2016, Ecological Modelling
      Citation Excerpt :

      Note that to ensure the coexistence between species spectra, a random coupling between predator and prey species had to be introduced (Hartvig et al., 2011), or the inclusion of schooling that prevents small species to be fully depleted by large predator species (Maury and Poggiale, 2013). The definition of additional traits such as habitat traits may be the next step to model species-diversified communities maintaining the coexistence (Hartvig and Andersen, 2013). Finally, some approaches go beyond generic trait-based representations by parameterizing explicitly species spectra to represent a particular set of species in a given ecosystem (see Blanchard et al., 2014; Spence et al., 2015 or the model Mizer, Scott et al. (2014)).

    View all citing articles on Scopus
    View full text