Biodiversity of plants: Broad-scale patterns and mechanisms

Abstract

One of the most obvious patterns in ecology is the geographic variation in species richness over broad spatial scales. However, despite observations of these patterns of richness for two centuries, mechanisms for the patterns remain controversial. This thesis is a study of the broad-scale biodiversity patterns of plants and the potential mechanisms behind those patterns. Richness-climate relationships often account for >80% of the spatial variance in richness. However, it has been suggested that richness-climate relationships differ significantly among geographic regions, and that there is no globally consistent relationship. Since there is little point in arguing about mechanisms before the patterns they predict have been documented, I investigated the global patterns of species and family richness of angiosperms in relation to climate. One hypothesized mechanism that predicts strong and consistent richness-climate patterns is the Physiological Tolerances hypothesis. This hypothesis predicts that the number of species found in a given location is a function of the number that can tolerate the climatic conditions in that place. Using global angiosperm family richness distributions and climatic data to estimate the climatic tolerances of plant families, I examined the relationships of potential family richness patterns to both climatic gradients and observed patterns of family richness. Observed family richness was not strongly related to potential family richness. However, both richness and potential richness were related to climate, but in different ways. Potential family richness necessarily sets an upper limit on observed richness levels, but observed richness did not generally reach that limit. In other words, there are generally many more species that can tolerate the climate in a particular area than actually occur there. A second hypothetical mechanism predicting strong consistent richness-climate patterns, the Energy-Diversity or Species-Energy hypothesis, suggests that for a species to persist in an area, there must be sufficient energy available in that area to support enough individuals of that species to maintain a viable population. Thus species richness will depend upon the number of individuals, which in turn depends upon energy availability. Using tree counts from 15 forest sites along a latitudinal gradient running from James Bay, Canada to Costa Rica, I tested whether a series of direct correlative links existed from climate to tree density to tree species richness. Despite a positive (but not statistically significant) correlation between AET and richness, the tree density per site was negatively correlated with energy. (Abstract shortened by UMI.)