The processes underlying continental-scale biodiversity patterns - Mechanisms at the interface of ecology and biogeography
Onderliggende processen van continentale biodiversiteitspatronen - mechanismen op het raakvlak van ecologie en biogeografie
Buschke, Falko; R0290164
For more than a century, naturalists have wondered why species occur where they do. Yet despite this persistent attention, identifying the mechanisms underlying contemporary biodiversity patterns remains a challenge and, as a consequence, a general theory of biodiversity is not forthcoming. The research presented in this thesis aimed at examining the importance of deterministic environment-driven processes in shaping the geographical distributions of individual vertebrate species and the aggregated properties of multi-species assemblages. To do this my co-authors and I used a macroecological approach to scrutinise diversity patterns in sub-Saharan Africa and applied a suite of mechanistic and statistical modelling techniques to distil evidence for deterministic causal processes. Furthermore, we also tested the ability of alternative mechanisms in explaining diversity patterns that have traditionally been attributed to climate mechanisms. Overall, we failed to find convincing evidence for environmental determinism at the continent-scale. We scrutinised an array of biological patterns and tested several alternative ecological and evolutionary mechanisms and all of our findings were within one of four categories. First, patterns consistent with environmental determinism - such as correlations between species' geographical ranges and climate gradients - arose even in the absence of environmental mechanisms. This weakens evidence for environmental determinism because we were unable to falsify alternative non-climate mechanisms such as continuous range expansion from a single point of origin. Second, evidence for environmental determinism could account for one biological pattern, but not for another equally relevant one. For example, we found that range expansion along continuous climate gradients could explain local, but not range-wide, patterns of species co-occurrence. This raises doubts around the ubiquity of climate-driven ecological mechanisms. Third, analytical tools used to identify environmental determinism at one spatial scale were inappropriate for use at a different scale. This is particularly relevant for macroecology, which regularly borrows analytical tools from smaller-scale ecological studies. As such, patterns could be misattributed to climate-driven ecological mechanisms in instances where analyses were carried out at unsuitable spatial scales. Lastly, in cases where non-climate mechanisms (specifically, the mid-domain effects hypothesis) have been tested and rejected, it is inappropriate to attribute these patterns to environmental determinism by default. Instead, it is equally likely that the original non-climate mechanism was poorly formulated with unrealistic assumptions.It was not our intention to discredit existing evidence for environmental determinism nor do we believe that the environment plays no role in shaping the distribution of life on earth. Instead, this study demonstrates the fragility of current macroecological knowledge and suggests reasons why strong evidence for environmental determinism is so hardto obtain. Ultimately, our findings allowed us to formulate several strategies for strengthening our understanding of ecological and evolutional processes at large spatial scales to, hopefully, build towards a general global theory of biodiversity.