Decomposing predictability to identify dominant causal drivers in complex ecosystems

Abstract

Ecosystems are complex systems of various physical, biological, and chemical processes. Since ecosystem dynamics are composed of a mixture of different levels of stochasticity and nonlinearity, handling these data is a challenge for existing methods of time-series based causal inferences. Here we show that, by harnessing contemporary machine learning approaches, the concept of Granger causality can be effectively extended to the analysis of complex ecosystem time series and bridge the gap between dynamical and statistical approaches. The central idea is to use an ensemble of fast and highly predictive artificial neural networks to select a minimal set of variables that maximizes the prediction of a given variable. It enables decomposition of the relationship among variables through quantifying the contribution of an individual variable to the overall predictive performance. We show how our approach, EcohNet, can improve interaction network inference for a mesocosm experiment and simulated ecosystems. The application of the method to a long-term lake monitoring dataset yielded new but interpretable results on the drivers causing cyanobacteria blooms, which is a serious threat to ecological integrity and ecosystem services. Since performance of EcohNet is enhanced by its predictive capabilities, it also provides an optimized forecasting of overall components in ecosystems. EcohNet could be used to analyze complex and hybrid multivariate time series in many scientific areas not limited to ecosystems. Significance StatementEffective use of ecosystem monitoring data to resolve global environmental issues is a major challenge of the 21st century ecology. A promising solution to address this challenge is a time-series-based causal analysis which can provide insight on the mechanical links between ecosystem components. In this work, a model-free framework named EcohNet is proposed. EcohNet utilizes ensemble predictions of echo state networks, which are known to be fast, accurate, and highly relevant for a variety of dynamical systems, and can robustly predict causal networks of ecosystem components. It also can provide an optimized forecasting of overall ecosystem components, and could be used to analyze complex and hybrid multivariate time series in many scientific areas, not limited to ecosystems.