Importance of Spatial and Temporal Dynamics in Species Regional Abundance and Distribution

Open Access
Publisher Website
Google Scholar
Abstract
Several recent models have been proposed to explain the pattern of species distribution within a region. Brown's niche—based model, Levins' stochastic immigration—extinction model, and Hanski's core—satellite hypothesis (a modification of Levins' model) differ in relative spatial scales, and degree of stochastic variation in species distribution over time. To test the predictions of these models, we analyzed the regional distribution and temporal dynamics of plant species across 19 sites in tallgrass prairie over an 8—yr period. Spatial patterns of species regional distribution were bimodal each year, a pattern in accord with the core—satellite hypothesis. In addition, temporal patterns of species distribution within a site were also bimodal. The dynamics of the core—satellite hypothesis are driven by patterns of immigration and extinction. Extinction was positively related to richness, however, in contrast to the theory of island biogeography, immigration was not related to species richness. In addition, the variances in the differences in species frequencies from one year to the next were greater than three times the mean, a condition necessary for bimodality in the core—satellite model, for 88 of 135 (65.1%) species. These results provide strong support in favor of the core—satellite hypothesis and imply that stochastic factors may be the primary determinant of plant species dynamics and plant community structure in grasslands at the scale of several kilometres. Evidence from prairie communities at much larger spatial scales supported the patterns predicted by Brown's niche—based model. Thus, these apparently competing models are, in fact, compatible when viewed at the appropriate spatial scale.

This publication has 0 references indexed in Scilit: