Biodiversity in a rapidly changing world: from local interactions to large scale patterns

Invasions and extinctions have reorganized the earth’s biota and altered biodiversity across all spatial scales. At the local scale, invasions have outpaced extinctions for many taxonomic groups. This suggests that food webs, which represent feeding interactions at the local scale, may be increasing in species richness. Importantly, the non-random addition and deletion of species has also altered the compositional similarity between regions or locales (beta diversity). The result is that spatially distinct assemblages have become more or less similar in species composition and abundance through the processes of biotic homogenization and biotic differentiation, respectively. In this dissertation, I addressed local scale interactions by exploring the influence of food web structure on invasion success in model food webs. I also quantified patterns of change in taxonomic and functional similarity across space and time to understand the effects of invasions and extinctions on large scale spatial patterns of diversity. I used a Lotka-Volterra food web model to develop predictions about how trophic structure influences invasion success (Chapter 1). I found that successful establishment in model food webs largely depends on the trophic level of the invader, due to interactions with adjacent trophic levels. My model makes four predictions that can be tested in natural or experimental communities; 1) invasion success of top predators will increase with greater diversity in native prey items, 2) basal invasion will be controlled by the number of native consumers, 3) invasive omnivore establishment will be controlled by diversity in the lowest trophic level of potential prey items, and 4) intermediate invasion success will be controlled by the diversity of native predators. I developed two methods that measure large scale spatial patterns of biodiversity. The dendrogram-based method, which quantifies change in taxonomic similarity, (Chapter 2) introduces three metrics that each describes a different aspect of change in taxonomic similarity as depicted by a dendrogram. This method is unique in that the spatial and historical affinities of assemblages are tracked through time providing insight into how evolutionary history and spatial dynamics influence patterns of homogenization. The utility of the dendrogram-based method was exemplified by the case study of the Hawaiian Island avifauna, which showed that between-island similarity in the historical time period follows the geologic history of the islands and the influence of prior extinction filters on the perceived homogenization of assemblages. The second method is a trait based method for quantifying change in functional similarity through time (Chapter 3). Simulations indicate that functional and taxonomic similarity are positively correlated as trait complementarity increases. Functional and taxonomic similarity are positively correlated for the breeding and foraging traits in bird assemblages at ten locales across the United States from 1968 to 2008. This relationship suggests a high level of trait complementarity among the breeding bird assemblages, but further empirical examples are necessary to determine the bounds of trait complementarity in real assemblages. The impact of humans on biodiversity is complex in that it involves measuring both taxonomic and functional attributes of communities across different spatial scales. Methods for elucidating anthropogenic impacts on biodiversity and ecosystem function must take this into consideration when assessing impacts and developing conservation planning.Ph.D.Includes bibliographical referencesby Benjamin Baise