Elsevier

Ecological Modelling

Volume 443, 1 March 2021, 109458
Ecological Modelling

Sociability interacts with temporal environmental variation to spatially structure metapopulations: A fish dispersal simulation in an ephemeral landscape

https://doi.org/10.1016/j.ecolmodel.2021.109458Get rights and content

Highlights

  • IBM simulating individual sociability's influence on metapopulation dispersal dynamics.

  • Individual personalities significantly structure metapopulations in space and time.

  • Dispersal and density, not landscape occupation, altered by fish personalities.

  • The effects of individual personalities can be overridden by environmental effects under some conditions.

Abstract

Metapopulation structure emerges from the dispersal of individuals among spatially distinct patches across a low-quality matrix. While dispersing agents are typically modeled as functionally identical with limited, linear and non-directional dispersal, field studies argue for the incorporation of intraspecific trait variation, such as behavioral, into the modeling of dispersal. Individual personality is of growing interest as a trait that affects dispersal metrics yet remains understudied at large, field-relevant spatial scales. Here, we used an individual-based model to investigate the influence of sociability, a personality-type known to be linked to density-dependent dispersal decisions, on a metapopulation at the individual, local and regional scales. As personality dependent dispersal is temporally dependent, we also examined how emergent properties of the metapopulation varied across two dispersal scenarios: long and short dispersal windows. Overall, our results support the growing evidence that individual personalities can significantly structure metapopulations in both space and time, through negative relationships between sociability and dispersal distance and positive relationships between sociability and patch density. However, we also demonstrate the effects of personality are limited by environmental factors. At the larger spatial scales of our model, patch density as a function of distance from source, and proportion of the landscape occupied were more strongly influenced by the dispersal window than by the personality. Using these results, we derive hypotheses to be examined in future empirical research. As the climate continues to shift, and dispersal models are updated to reflect ever changing conditions, these results suggest individual personality types and the personality type by environment interaction should be considered to accurately reflect dispersal and manage valuable metasystems.

Introduction

Metapopulation structure emerges from the dispersal of propagules among spatially distinct patches across a low-quality matrix, determining metapopulation stability and persistence (Hanski and Gilpin, 1991; Pulliam et al., 1992). The rate at which these propagules disperse (Gotelli, 1991), the pathways they use (Altermatt et al., 2011; Baguette and Van Dyck, 2007), and their use of environmental information while dispersing (Proulx et al., 2013) have been foci of research on the spatial structuring of metapopulations. Yet, a key assumption of much of this work is that individual propagules are functional equivalents within the metapopulation and thus the dispersal of propagules has been assumed to be limited, linear, non-directional and importantly, traditionally modeled as a single parameter with no variance (Hanski and Gilpin 1991; Wang and Altermatt 2019, but see Bani et al., 2019). However, it is now clear that intraspecific trait variation, particularly for dispersal associated physiological, morphological and behavioral traits, can influence emergent metapopulation structure (Hawkes, 2009; Sutherland et al., 2014). Among these traits, personality-types (PT) are now recognized as particularly influential dispersal-related traits affecting the spatial structuring of metapopulations (Conrad et al., 2011; Dhellemmes et al., 2020; Spiegel et al., 2017).

A PT, also referred to as behavioral-type, is defined as the within and between individual consistent behavior, observed across time and/or ecological context in response to a given stimulus (Sih et al., 2012). PT have been shown to influence key metapopulation dynamics, including propagule dispersal tendencies (Cote et al., 2010a; Duckworth and Badyaev, 2007; Myles-Gonzalez et al., 2015), speeds (Maes et al., 2013), distances (Cote et al., 2010b), competitive abilities (Capelle et al., 2015; Groen et al., 2012), physiological states (Myles-Gonzalez et al., 2015), parental care (Duckworth and Badyaev, 2007), home range size (Schirmer et al., 2019) and colonization behaviors (Duckworth, 2008; Groen et al., 2012). In order to influence metapopulation structure through dispersal, a PT must, at minimum, affect both the spatial (e.g., distance) and temporal (e.g., rate) components of dispersal, at ecologically relevant scales. Among other behaviors that characterize an individual's PT, ‘sociability’ or the propensity of an individual to associate with conspecifics (Cote and Clobert, 2007), has been suggested to influence both of these aspects of dispersal by mediating emigration and immigration probabilities, as well as dispersal rates and distances (Cote et al., 2010a; Cote et al., 2010b), and thus should be a strong determinant of metacommunity dynamics, yet its role remains understudied.

Sociability's potential influence on metapopulation structure stems from how local density-dependence at the patch scale interacts with an individual's sociability to influence dispersal decisions. As conceptualized and reviewed by Cote et al. (2010a), asocial individuals (i.e., those who prefer low conspecific densities), disperse away from high density patches in search of preferable conditions. Patch density decreases as they disperse, encouraging emigration of social individuals, who seek higher density conditions. Through multiple iterations of this sequential, stepwise dispersal process, high- and low-density patches are established and go extinct over time and space. The emergent structure of the metapopulation then becomes dependent on the members’ sociability distribution (Cote et al., 2010a; Rehage et al., 2016). Some evidence exists for this, at small spatial scales. Cote et al. (2010b) showed a negative relationship between sociability and individual dispersal distances in a fish mesocosm, with asocial individuals having the largest dispersal distances. Similarly, in a reptile macrocosm, social individuals tended to disperse away from low density patches more readily than those starting in high density areas, while the reverse was true of less social individuals (Cote and Clobert 2007). However, there is still a lack of information regarding how this translates to large spatial scales (10s-100 s of km) and field settings, and how sociability interacts with temporally varying environmental conditions that may influence dispersal. Ancillary evidence at larger scales exists, such as the tendency of burrowing owls to disperse further from low conspecific density patches relative to high density patches (Moiron et al., 2020), and more bold, likely asocial, bank voles tend to move the most (Schirmer et al., 2019), but no field-scale assessments of the role of sociability have been conducted nor modeled.

In this study, we ask how does personality dependent dispersal (PDD), particularly sociability, influence metapopulation structure in both space and time. To do this, we built a behavioral individual-based model (IBM), parameterized with information from a real-world metapopulation. Using this IBM, we experimentally manipulated the sociability of individuals to test hypotheses on how variation in the personality composition of the metapopulation affected the dispersal of individuals, and both local (patch scale), and regional (across patches) attributes of the metapopulation (Table 1). The premise of our model is that personality interacts with patch density (as per Cote et al. 2010a’s hypothesis), both responding to and influencing it, to determine emigration out of patches and thus affect metapopulation structure. The goal of our model is to examine the consequences of this interaction of personality and patch density to determine individual, patch and regional metapopulation attributes. We hypothesized that the distance individuals traveled would be negatively related to their sociability, as per Cote et al. 2010b. At the patch scale, patch sociability was expected to positively correlate with patch density as asocial individuals would actively seek to establish low density patches while social individuals would actively seek to congregate in high-density patches. Finally, at the regional scale, we expected to observe a negative relationship between population sociability and the proportion of the landscape occupied, as individuals in asocially-skewed populations were expected to travel further from the source to achieve low densities, and thus colonize more of the landscape.

Understanding the effect of PDD on metapopulation structure is a particularly pressing need in aquatic systems. While many aquatic habitats are temporally stable with functionally unlimited dispersal windows, or periods of time suitable for dispersal and colonization of new habitat patches, dispersal can be limited in other ephemeral landscapes. For instance, in rainfall-driven pulsing systems, the extent of marsh inundation and thus habitat connectivity among habitat patches varies seasonally and yearly with drought severity (Davis et al., 2017). Davis et al. (2017), in an analysis of an ephemeral wetland metacommunity, showed that precipitation and the presence of fishes closely tied to the rain-induced flood levels, were strong predictors of amphibian colonization during a multiyear period of severe climatic variability. Thus, in addition to testing hypotheses about the role of PDD on individual dispersal, inter and intra-patch dynamics, we tested a long vs. a short dispersal window scenario (Table 1). At the individual level, we hypothesize that a shorter dispersal window would lessen differences in the dispersal behavior of social vs. asocial individuals resulting in a weaker relationship between individual sociability and distance traveled. Patch density was used both for model calibration and for testing a priori hypotheses. As the mechanism being investigated enforces a positive relationship between patch density and sociability, this pattern was used for model calibration. How this pattern changes with dispersal windows, however, was not an enforced pattern. With that said, we did not anticipate a significant effect of dispersal window length on the association between population sociability and patch density. At the regional scale we anticipated a weakening of the relationship between population sociability and the proportion of the landscape occupied at longer dispersal windows, as these would allow for full exploration and thus possible saturation of the model environment, resulting in all patches being occupied regardless of sociability.

Section snippets

Model overview

Our IBM was built to simulate the structuring of a metapopulation as a function of intraspecific variation in sociability and two dispersal window scenarios, using NetLogo v. 5.3.1 (Wilensky, 1999). The model simulated the departure of fish out of a source population, their movement across a habitat matrix in search of suitable patches, and their settlement upon encountering a suitable patch. We varied the presence and skewness of sociability in this population of dispersing propagules in order

Individual scale

As predicted, socially skewed populations traveled less across the model environment than asocials. The distance traveled by agents was negatively related to their sociability under both long (F = 3,265,001, t= −1807, R2adj.=0.45, P<0.001), and short dispersal windows (F = 170,370, t= −412.8, R2adj.=0.04, P<0.001), but the slope was significantly steeper for the longer dispersal window (F = 2,599,566, P<0.001), indicating how shorter dispersal windows limit dispersal and the effect of

Discussion

The mechanisms structuring metapopulations have been of central interest to ecologists for several decades (e.g., Harrison et al., 1988, Gotelli 1991, Govindan et al., 2015, Deans and Chalcraft 2017). While environmental factors and interspecific interactions are known to be influential (Bond et al., 2015; Pringle, 2003), individual personalities can also significantly contribute to the structuring of metapopulations through personality-dependent dispersal (PDD) (e.g., Cote and Clobert 2007,

Conclusions

Overall, our results support the growing evidence that individual personalities are particularly important in the field of metasystem ecology, as they influence dispersal rates and density-dependent processes. In agreement with past studies, we demonstrate here that individual personalities can influence metapopulation properties. It follows that similar effects exist at other metasystem scales (i.e. metacommunity and metaecosystem) as well. Future works should seek to evaluate the hypotheses

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to thank Florida International University for financial assistance during the conceptualization and development of this model, in the form of a teaching assistantship and post-doctoral fellowship. The feedback provided by Dr. Volker Grimm and Dr. Steven Railsback during their summer agent-based modeling workshop was instrumental in code development for this manuscript. Additionally, the peer-review service provided by the editor and external reviewers associated with this

References (82)

  • S. Wang et al.

    Metapopulations revisited: the area-dependence of dispersal matters

    Ecology

    (2019)
  • P.B. Adler et al.

    A niche for neutrality

    Ecol. Lett.

    (2007)
  • Alonso D., Etienne R., McKane A. The merits of neutral theory. Trends in Ecology and Evolution. 2006....
  • F. Altermatt et al.

    Effects of connectivity and recurrent local disturbances on community structure and population density in experimental metacommunities

    PLoS ONE

    (2011)
  • M. Baguette et al.

    Landscape connectivity and animal behavior: functional grain as a key determinant for dispersal

    Landsc. Ecol.

    (2007)
  • R. Bani et al.

    Dispersal traits interact with dynamic connectivity to affect metapopulation growth and stability

    Theor. Ecol.

    (2019)
  • J.R. Blanchard

    A Confluence of Invasion, Behavioral, and Theoretical Ecology: What Drives Ephemeral Metacommunity Re-Assembly?

    (2018)
  • J.R. Blanchard

    Sociability and Time Can Restructure Metapopulations

    (2021)
  • N.R. Bond et al.

    Fish population persistence in hydrologically variable landscapes

    Ecol. Appl.

    (2015)
  • D. Bonte et al.

    Costs of dispersal

    Biol. Rev.

    (2012)
  • D.J. Booth

    Juvenile groups in a coral-reef damselfish: density-dependent effects on individual fitness and population demography

    Ecology

    (1995)
  • M.S. Botham et al.

    Predator choice in the field; grouping guppies, Poecilia reticulata, receive more attacks

    Behav. Ecol. Sociobiol.

    (2005)
  • T. Brodin et al.

    Personality-dependent survival of the invasive mosquitofish: being social can be deadly

    Aquat. Invasions

    (2019)
  • J.W. Brownscombe et al.

    Range expansion dynamics of the invasive round goby (Neogobius melanostomus) in a river system

    Aquat. Ecol.

    (2012)
  • Capelle, P.M., Mccallum, E.S., Balshine, S., 2015. Aggression and sociality: conflicting or complementary traits of a...
  • J. Clobert et al.

    Informed dispersal, heterogeneity in animal dispersal syndromes and the dynamics of spatially structured populations

    Ecol. Lett.

    (2009)
  • J.L. Conrad et al.

    Behavioural syndromes in fishes: a review with implications for ecology and fisheries management

    J. Fish Biol.

    (2011)
  • Core R. A Language and environment for statistical computing. R Foundation for statistical computing, Vienna, Austria....
  • J. Cote et al.

    Social personalities influence natal dispersal in a lizard

    Proc. Biol. Sci.

    (2007)
  • J. Cote et al.

    Personality-dependent dispersal: characterization, ontogeny and consequences for spatially structured populations

    Philos. Trans. R. Soc. Lond. B. Biol. Sci.

    (2010)
  • J. Cote et al.

    Personality-dependent dispersal in the invasive mosquitofish: group composition matters

    Proc. R. Soc. Biol. Sci.

    (2011)
  • Julien Cote et al.

    Personality traits and dispersal tendency in the invasive mosquitofish (Gambusia affinis)

    Proc. R. Soc. Biol. Sci.

    (2010)
  • J. Cote et al.

    Personality-dependent dispersal cancelled under predation risk

    Proceedings of the Royal Society B, Biological Sciences

    (2013)
  • S.M. Davis et al.
    (2005)
  • C.L. Davis et al.

    Species interactions and the effects of climate variability on a wetland amphibian metacommunity

    Ecol. Appl.

    (2017)
  • A. Deacon et al.

    How reproductive ecology contributes to the spread of a globally invasive fish

    PloS one.

    (2011)
  • R.A. Deans et al.

    Matrix context and patch quality jointly determine diversity in a landscape-scale experiment

    Oikos

    (2017)
  • B. Diaz Pauli et al.

    Behavioural responses to human-induced change: why fishing should not be ignored

    Evol. Appl.

    (2017)
  • C.J. Dibble et al.

    Intraspecific priority effects and disease interact to alter population growth

    Ecology

    (2014)
  • R.A. Duckworth

    Adaptive dispersal strategies and the dynamics of a range expansion

    Am. Nat.

    (2008)
  • R.A. Duckworth et al.

    Coupling of dispersal and aggression facilitates the rapid range expansion of a passerine bird

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

    (2007)
  • Cited by (3)

    View full text