Patterns of co-occurrences in a killifish metacommunity are more related with body size than with species identity

Authors

  • Andrés Canavero,

    Corresponding author
    1. Departamento de Ecología, Facultad de Ciencias Biológicas, Pontificia Universidad Católica de Chile, Santiago, CP, Chile
    2. Centro Universitario de Rivera (CUR), Universidad de la República, Rivera, Uruguay
    • Corresponding author.

    Search for more papers by this author
  • Daniel Hernández,

    1. Departamento de Ecología y Evolución, Facultad de Ciencias & Centro Universitario Regional Este (CURE), Universidad de la República, Montevideo, Uruguay
    Search for more papers by this author
  • Matías Zarucki,

    1. Departamento de Ecología y Evolución, Facultad de Ciencias & Centro Universitario Regional Este (CURE), Universidad de la República, Montevideo, Uruguay
    Search for more papers by this author
  • Matías Arim

    1. Departamento de Ecología y Evolución, Facultad de Ciencias & Centro Universitario Regional Este (CURE), Universidad de la República, Montevideo, Uruguay
    Search for more papers by this author

Abstract

Body size may be more important than species identity in determining species interactions and community structure. However, co-occurrence of organisms has commonly been analysed from a taxonomic perspective and the body size is rarely taken into account. On six sampling occasions, we analysed patterns of killifish co-occurrences in nestedness (tendency for less rich communities to be subsamples of the richest), checkerboard structure (tendency for species segregation), and modularity (tendency for groups to co-occur more frequently than random expectation) in a pond metacommunity located in Uruguay. We contrasted co-occurrence patterns among species and body size-classes (individuals from different species were combined into size categories). The analysis was performed at two spatial scales: ponds (communities) and sample units within ponds. Observed nestedness was frequently smaller than the null expectation, with significantly greater deviations for body size-classes than for species, and for sample units than for communities. At the sample unit level, individuals tended to segregate (i.e. clump into a checkerboard pattern) to a larger extent by body size rather than by taxonomy. Modularity was rarely detected, but nevertheless indicated a level of taxonomic organization not evident in nestedness or checkerboard indices. Identification of the spatial scale and organization at which ecological forces determine community structure is a basic requirement for advancement of robust theory. In our study system, these ecological forces probably structured the community by body sizes of interacting organisms rather than by species identities.

Ancillary