Community assembly on isolated islands: macroecology meets evolution

A. J. Rominger, K. R. Goodman, J. Y. Lim, E. E. Armstrong, L. E. Becking, G. M. Bennett, M. S. Brewer, D. D. Cotoras, C. P. Ewing, J. Harte, N. D. Martinez, P. M. O'Grady, D. M. Percy, D. K. Price, G. K. Roderick, K. L. Shaw, F. S. Valdovinos, D. S. Gruner, R. G. Gillespie

Research output: Contribution to journalArticlepeer-review

47 Scopus citations


Aim: Understanding how ecological and evolutionary processes together determine patterns of biodiversity remains a central aim in biology. Guided by ecological theory, we use data from multiple arthropod lineages across the Hawaiian archipelago to explore the interplay between ecological (population dynamics, dispersal, trophic interactions) and evolutionary (genetic structuring, adaptation, speciation, extinction) processes. Our goal is to show how communities develop from the dynamic feedbacks that operate at different temporal and spatial scales. Location: The Hawaiian islands (19–22° N, 155–160° W). Methods: We synthesize genetic data from selected arthropods across the Hawaiian archipelago to determine the relative role of dispersal and in situ differentiation across the island chronosequence. From four sites on three high islands with geological ages ranging from < 1 Ma to 5 Ma, we also generate ecological metrics on plant–herbivore bipartite networks drawn from the literature. We compare the structure of these networks with predictions derived from the principle of maximum information entropy. Results: From the perspective of the island chronosequence we show that species at lower trophic levels develop population genetic structure at smaller temporal and spatial scales than species at higher trophic levels. Network nestedness decreases while modularity increases with habitat age. Single-island endemics exhibit more specialization than broadly distributed species, but both show the least specialization in communities on middle-aged substrates. Plant–herbivore networks also show the least deviation from theoretical predictions in middle-aged communities. Main conclusions: The application of ecological theory to island chronosequences can illuminate feedbacks between ecological and evolutionary processes in community assembly. We show how patterns of population genetic structure, decreasing network nestedness, increasing network modularity and increased specialization shift from early assembly driven by immigration, to in situ diversification after > 1 Myr. Herbivore–plant communities only transiently achieve statistical steady state during assembly, presumably due to incomplete assembly from dispersal in the early stages, and the increasing influence of island ontogeny on older islands.

Original languageEnglish (US)
Pages (from-to)769-780
Number of pages12
JournalGlobal Ecology and Biogeography
Issue number7
StatePublished - Jul 1 2016


  • Arthropods
  • Hawaii
  • chronosequence
  • maximum entropy
  • networks
  • population genetics

ASJC Scopus subject areas

  • Global and Planetary Change
  • Ecology, Evolution, Behavior and Systematics
  • Ecology


Dive into the research topics of 'Community assembly on isolated islands: macroecology meets evolution'. Together they form a unique fingerprint.

Cite this