COS 75-8 - A simulation-based approach to modern coexistence theory for empirical applications: Non-spatial mechanisms

Wednesday, August 9, 2017: 10:30 AM
B116, Oregon Convention Center
Stephen P. Ellner, Ecology and Evolutionary Biology, Cornell University, Ithaca, NY, Robin E. Snyder, Biology, Case Western Reserve University, Cleveland, OH, Peter B. Adler, Department of Wildland Resources and the Ecology Center, Utah State University, Logan, UT and Giles Hooker, Biological Statistics and Computational Biology, Cornell University, Ithaca, NY

Chesson (2000) showed how many previous studies on specific coexistence mechanisms, such as resource partitioning or temporal niche partitioning, could be unified into a general framework often called Modern Coexistence Theory (MCT). MCT has inspired new basic research about the mechanisms maintaining species coexistence, with important implications for understanding how high-diversity communities persist and for predicting impacts of environmental changes such as habitat fragmentation, nutrient enrichment, and changes in climate variability. However, MCT has limitations that have restricted empirical applications: new study systems often require a new model and a new, technically complex mathematical analysis; the standard categorization of coexistence mechanisms is not detailed or flexible enough for the possible coexistence mechanisms in many communities; and a key assumption of the underlying mathematical theory fails to hold in many data-driven models.


We present first steps towards a general simulation-based approach to MCT that will remove these obstacles. The key ingredients are (i) a general method for partitioning population growth rates into contributions from different mechanisms and their interactions, and (ii) evaluating the contributions through counterfactual simulations rather than analytic small-variance approximations. We apply this new approach to an experimental study by Descamps-Julien and Gonzalez (2005) on diatom species coexistence under periodic variation in temperature. Our approach diagnoses that persistence depends on fluctuation-dependent mechanisms, as was shown experimentally. For one of the two diatom species, the dominant fluctuation-dependent contributions are terms that are absent from MCT’s standard partitioning of non-spatial mechanisms: relative nonlinearity in response to temperature (analogous to, but distinct from, relative nonlinearity of competition), and the interaction between the effects of variance in competition and in temperature.