Trait-based approaches purport to create a more general and predictable framework for understanding species distributions, but these models are inadequately unvetted, despite their use in large systems. Trait-based approaches are also used to understand various aspects of a local environment, such as nutrient concentrations in aquatic systems. For environmental outcomes, response traits, traits of individual organisms that affect their environment, are the traits of interest. We test if response traits of focal organisms can predict environmental level outcomes in small-scale, well controlled systems, in which all parameters are known. Specifically, we use response traits of cladocerans Daphnia magna and Ceriodaphnia rigaudi to predict algal and ammonium changes in a closed aquatic system. We measured response traits of lab stocks of these zooplankton species, including ingestion, ammonium excretion, and reproduction rates, as well as the growth rate of Daphnia magna at different algal concentrations in our laboratory conditions. We also measured our primary algal genus’ (Scenedesmus sp.) growth rates at different ammonium concentrations. Using these values we parameterize a series of differential equations to estimate the population dynamics of both zooplankton and algae, as well as nutrient cycling. We estimated how the model matches an experimental (six weeks) population dynamics and ammonium concentrations, in Daphnia magna with algae, Ceriodaphnia rigaudi with algae, and both cladocerans together.
Results/Conclusions:
As part of evaluating the effectiveness of using species traits as a foundation for prediction, we included a disturbance treatment (a single influx of fertilizer) to see if our dynamic models are able to predict system responses to novel conditions. Preliminary results suggest that even in our undisturbed treatments, stochastic events such as novel invasions or shifts in dominant functional species groups limited our ability to make accurate predictions in many replicates. Even though our systems are small and thus highly susceptible to small disturbances compared to lakes or oceans, these results suggest that using response traits to predict system level changes will be difficult. Take home message: The difficulty in making process predictions using traits may have a deeper foundational problem arising from the mismatch between the fluid nature of species assemblage composition and the less flexible expectations of the role of traits.