Document: KAR-3-82-27

Trophic and taxonomic patterns of impact of invasive species: A meta-analysis.

GOODELL, K.* ¹, M.J.WONHAM ², B.VON HOLLE ³ and I.M.PARKER ⁴

State University of New York, Stony Brook, NY 11794 USA ¹
University of Washington, Seattle, WA, USA ²
University of Tennessee, Knoxville, TN, USA ³
University of California, Santa Cruz, CA, USA ⁴

Abstract:
The ecological impact of successful invaders by introduced species depends, in part, on characteristics of the invader and its interaction with the invaded community. We conducted a meta-analysis to examine patterns and test ecological predictions of invader impacts based on characteristics of the particular invader and of the invasion pattern. We quantitatively synthesized results from 90 field studies that provided 618 different measures of impact of plant, insect, fish, or marine invertebrate invaders. We divided response variables into impacts at the individual, population, community or ecosystem levels, expressed as the response ratio (lnRR). Invader taxon, trophic level, taxonomic relationship, and type of interaction between invader and response species all affected patterns of impact. At the population level, effects of insect and marine invertebrate invaders exceeded those of fish or plants, but we found no differences in effects at other levels. For insects (population impacts) and fish (community impacts), omnivores consistently had larger effects than predators or herbivores. For fish (population level) and insects (community level), indirect invader effects exceeded competitive effects; predation effects were intermediate. The impacts of plant invaders did not differ significantly between competitors and invaders that altered habitat quality. Competitive impacts of invading insects were higher for response species related at the family level or closer, but this pattern did not hold for other taxa. Invasion scale and pathway also affected impacts. Contiguous range extensions showed lower impacts than disjunct invasions at the population level for all taxa together and for fish. For all taxa together, accidental introductions had significantly higher impacts than intentional introductions at the population, community and individual level. Insects intentionally introduced for biocontrol had lower population level effects (on non-target organisms) than accidentally introduced insects. We developed a randomization procedure to correct for the effect of non-independence among effects measured within a single study. Using this technique, fewer comparisons showed significant heterogeneity.

Keywords: alien species, competition, herbivory, introduced species, invasions, meta-analysis, omnivory, predation, trophic interactions

This abstract is being presented at: 11:30 AM in session:
Oral Session #56: Metapopulation Analysis.