Neighbour tolerance,not suppression,provides competitive advantage to non‐native plants |
| |
Authors: | Marina Golivets Kimberly F. Wallin |
| |
Affiliation: | 1. The Rubenstein School of Environment and Natural Resources, The University of Vermont, Burlington, VT, USA;2. USDA Forest Service, Northern Research Station, Burlington, VT, USA |
| |
Abstract: | High competitive ability has often been invoked as a key determinant of invasion success and ecological impacts of non‐native plants. Yet our understanding of the strategies that non‐natives use to gain competitive dominance remains limited. Particularly, it remains unknown whether the two non‐mutually exclusive competitive strategies, neighbour suppression and neighbour tolerance, are equally important for the competitive advantage of non‐native plants. Here, we analyse data from 192 peer‐reviewed studies on pairwise plant competition within a Bayesian multilevel meta‐analytic framework and show that non‐native plants outperform their native counterparts due to high tolerance of competition, as opposed to strong suppressive ability. Competitive tolerance ability of non‐native plants was driven by neighbour's origin and was expressed in response to a heterospecific native but not heterospecific non‐native neighbour. In contrast to natives, non‐native species were not more suppressed by hetero‐ vs. conspecific neighbours, which was partially due to higher intensity of intraspecific competition among non‐natives. Heterogeneity in the data was primarily associated with methodological differences among studies and not with phylogenetic relatedness among species. Altogether, our synthesis demonstrates that non‐native plants are competitively distinct from native plants and challenges the common notion that neighbour suppression is the primary strategy for plant invasion success. |
| |
Keywords: | Bayesian multilevel meta‐analysis competitive strategy inter‐ vs. intraspecific competition net neighbour effect non‐native invasive plants pairwise competition phylogenetic correction statistical non‐independence |
|
|