Ecological release from interspecific competition leads to decoupled changes in population and individual niche width

Abstract

A species's niche width reflects a balance between the diversifying effects of intraspecific competition and the constraining effects of interspecific competition. This balance shifts when a species from a competitive environment invades a depauperate habitat where interspecific competition is reduced. The resulting ecological release permits population niche expansion, via increased individual niche widths and/or increased among-individual variation. We report an experimental test of the theory of ecological release in three-spine stickleback (Gasterosteus aculeatus). We factorially manipulated the presence or absence of two interspecific competitors: juvenile cut-throat trout (Oncorhynchus clarki) and prickly sculpin (Cottus asper). Consistent with the classic niche variation hypothesis, release from trout competition increased stickleback population niche width via increased among-individual variation, while individual niche widths remained unchanged. In contrast, release from sculpin competition had no effect on population niche width, because increased individual niche widths were offset by decreased between-individual variation. Our results confirm that ecological release from interspecific competition can lead to increases in niche width, and that these changes can occur on behavioural time scales. Importantly, we find that changes in population niche width are decoupled from changes in the niche widths of individuals within the population.

Figure 1.

Illustration of three potential forms of ecological release. Consider a population that (a) initially coexists with an interspecific competitor; the population niche width is indicated by a thick curve, and niche widths of four individuals are indicated by shorter, thin lines. Release from competition can lead to (b) increased individual and population niche widths (parallel release); (c) increased individual but not population niche widths, because expansion is offset by decreased among-individual variation (individual release); or (d) increased population but not individual niche widths, via increased among-individual variation (niche variation hypothesis). Each of these scenarios (b–d) is plotted as (e) a vector in a niche space graph. In this type of graph, we plot individual versus population niche width (WIC versus TNW). Because TNW = WIC + BIC, any population must fall on or below the solid line where WIC/TNW = 1. Thin dashed lines below this represent increasing isoclines of individual specialization (smaller WIC/TNW). Ecological release can be plotted as a vector in this space, from the high to low competition niche widths.

Figure 2.

Effect of competitor removal on stickleback population total niche width (TNW). To visually represent the effect of release from trout, we calculated the average TNW with and without trout for a given block, averaging across sculpin treatments. Similarly, TNW with versus without sculpin are averaged across trout treatments within a block. For simplicity, we do not illustrate the non-significant sculpin × trout interaction. Lines connect competitor present versus absent results for a given block. Blocks are colour-coded to permit comparisons across (a) and (b), and with other figures.

Figure 3.

Effect of ecological release from (a) trout and (b) sculpin on stickleback population and individual niche widths. Release is plotted in a subset of the niche width space (explained in figure 1). Experimental effects are represented as vectors (one per experimental block). The vectors connect the mean (TNW, WIC) combination for competitor-present to competitor-absent treatments within a block, averaging across the other competitor treatments. As in figure 1, dotted lines represent isoclines of WIC/TNW, with individual specialization increasing as WIC/TNW declines from 1.0 towards zero. Blocks are colour-coded to correspond with other figures.

Figure 4.

Effect of competitor removal on the degree of individual specialization in stickleback (WIC/TNW). When WIC/TNW = 1, individuals have the same niche breadth as the population as a whole. As WIC/TNW gets smaller, individuals are increasingly specialized relative to their population and between-individual variation is proportionally larger. Each point is the average value for a given block of enclosures, averaging across the other competitor treatments. Lines connect competitor-present versus competitor-absent results for a given block. Blocks are also colour-coded to permit comparisons across (a) and (b) and with other figures. We do not illustrate the non-significant interaction between trout and sculpin removal.