Environmental and spatial controls of biotic assemblages in a discrete semi-terrestrial habitat: comparison of organisms with different dispersal abilities sampled in the same plots : Dispersal ability and environmental filtering in biotic assemblages

Emergence of the metacommunity concept has made a substantial contribution to better understanding of the community composition and dynamics in a regional context. However, long-term field data for testing of available metacommunity models are still scarce, and the extent to which these models apply to the real world remains unknown. Tests conducted so far have largely sought to fit data on the entire regional set of species to one of several metacommunity models, implicitly assuming that all species operate similarly over the same set of sites. However, species differ in their habitat use. These differences can, in the most general terms, be expressed as a gradient of habitat specialization (ranging from habitat specialists to habitat generalists). We postulate that such differences in habitat specialization will have implications for metacommunity dynamics. Specifically, we predict that specialists respond more to local processes and generalists respond to regional spatial processes. We tested these predictions using natural microcosm communities for which long-term (nine-year) environmental and population dynamics data were available. We used redundancy analysis to determine the proportion of variation explained by environmental and spatial factors. We repeated this analysis to explain variation in the entire regional set of species, in generalist species only, and in specialists only. We further used ANOVA to test for differences in the proportions of explained variation. We found that habitat specialists responded primarily to environmental factors and habitat generalists responded mainly to spatial factors. Thus, from the metacommunity perspective, the dynamics of habitat specialists are best explained by a combination of species sorting and mass effects, while that of habitat generalists are best explained by patch dynamics and neutral models. Consequently, we infer that a natural metacommunity can exhibit complicated dynamics, with some groups of species (e.g., habitat specialists) governed according to environmental processes and other groups (e.g., habitat generalists) governed mainly by dispersal processes.

This paper proposes a new way of using forward selection of explanatory variables in regression or canonical redundancy analysis. The classical forward selection method presents two problems: a highly inflated Type I error and an overestimation of the amount of explained variance. Correcting these problems will greatly improve the performance of this very useful method in ecological modeling. To prevent the first problem, we propose a two-step procedure. First, a global test using all explanatory variables is carried out. If, and only if, the global test is significant, one can proceed with forward selection. To prevent overestimation of the explained variance, the forward selection has to be carried out with two stopping criteria: (1) the usual alpha significance level and (2) the adjusted coefficient of multiple determination (Ra(2)) calculated using all explanatory variables. When forward selection identifies a variable that brings one or the other criterion over the fixed threshold, that variable is rejected, and the procedure is stopped. This improved method is validated by simulations involving univariate and multivariate response data. An ecological example is presented using data from the Bryce Canyon National Park, Utah, U.S.A.

Dispersal can limit the ranges of species and the diversity of communities. Despite its importance, little is known about its role in freshwater habitats and its relation to habitat type (lentic vs. lotic), especially for organisms with cryptic dispersal methods such as plankton. Poor dispersers are expected to show more clumped distributions or greater spatial autocorrelation (SA) in community composition than good dispersers. We examined patterns of SA across freshwater taxa with different dispersal modes (active vs. passive) and their association with habitat type (lake vs. stream) using 18 spatially explicit community composition data sets. We found significant relationships between SA and body size among taxa in lake habitats, but not in streams. However, the increase in SA with body size in lakes was driven entirely by fishes-organisms ranging in size from diatoms to macro-invertebrates showed equivalent levels of SA. These results support the idea that large organisms are less effective dispersers in aquatic environments, resulting in greater SA in community structure over broad scales. Streams may be effectively more connected than lakes as patterns of SA and body size were weaker in lotic habitats. Our data suggest that the critical threshold where greater body size increases dispersal limitation seems to come at the juncture between invertebrates and vertebrates rather than that between unicellular and multicellular organisms as has been previously suggested.