Interpreting random forest analysis of ecological models to move from prediction to explanation.

As modeling tools and approaches become more advanced, ecological models are becoming more complex. Traditional sensitivity analyses can struggle to identify the nonlinearities and interactions emergent from such complexity, especially across broad swaths of parameter space. This limits understanding of the ecological mechanisms underlying model behavior. Machine learning approaches are a potential answer to this issue, given their predictive ability when applied to complex large datasets. While perceptions that machine learning is a "black box" linger, we seek to illuminate its interpretive potential in ecological modeling. To do so, we detail our process of applying random forests to complex model dynamics to produce both high predictive accuracy and elucidate the ecological mechanisms driving our predictions. Specifically, we employ an empirically rooted ontogenetically stage-structured consumer-resource simulation model. Using simulation parameters as feature inputs and simulation output as dependent variables in our random forests, we extended feature analyses into a simple graphical analysis from which we reduced model behavior to three core ecological mechanisms. These ecological mechanisms reveal the complex interactions between internal plant demography and trophic allocation driving community dynamics while preserving the predictive accuracy achieved by our random forests.

A bioenergetic framework for aboveground terrestrial food webs.

Bioenergetic approaches have been greatly influential for understanding community functioning and stability and predicting effects of environmental changes on biodiversity. These approaches use allometric relationships to establish species' trophic interactions and consumption rates and have been successfully applied to aquatic ecosystems. Terrestrial ecosystems, where body mass is less predictive of plant-consumer interactions, present inherent challenges that these models have yet to meet. Here, we discuss the processes governing terrestrial plant-consumer interactions and develop a bioenergetic framework integrating those processes. Our framework integrates bioenergetics specific to terrestrial plants and their consumers within a food web approach while also considering mutualistic interactions. Such a framework is poised to advance our understanding of terrestrial food webs and to predict their responses to environmental changes.