Assistant Professor University of California Davis, United States
Background/Question/Methods The field of mutualistic networks has developed tremendously over the past 15+ years, but in nearly all of that development, theory and empirical advances have progressed largely separately. This lack of integration is perhaps unsurprising given the challenges in, for example, parameterizing species-rich population dynamic network models, where the number of parameters quickly becomes impractical to estimate in the field. In this talk we discuss 1) recent advances in data-theory integration in mutualistic networks; 2) ways that empiricists can collect data that allow for better use by theoreticians; and 3) ways that theoreticians can develop models that both include more empirical data as inputs and lead to model outputs that are more easily testable using empirical data. We also highlight collaborative ongoing work focused on closing the empirical-theory gap in pollination networks in particular.
Results/Conclusions Recent and ongoing work underscore several avenues for theory-data integration in mutualistic networks. Given the challenges in pairwise interaction parameterization, we discuss as one alternative the use of empirically validated functional forms (functional and numerical responses) in models. In most mutualistic network models, these currently come from first-principles models lacking empirical validation. We also discuss network data collection that allows for better data use by theoreticians, including careful delineation of both space and time in sampling, as well as estimating organismal abundances independently of interaction sampling. We synthesize some of the approaches that have been taken in the food webs literature that have generally led to better integration in that field. We will close by discussing a general roadmap for better integration of data and theory in mutualistic networks.