1. Although ecological models used to make predictions from underlying covariates have a record of success, they also suffer from limitations. They are typically unable to make predictions when the value of one or more covariates is missing during the testing. Missing values can be estimated but methods are often unreliable and can result in poor accuracy. Similarly, missing values during the training can hinder parameter estimation of many ecological models. Bayesian networks can handle these and other limiting issues, such as having highly correlated covariates.However, they are rarely used to their full potential.2. Indeed, Bayesian networks are commonly used to evaluate the knowledge of experts by constructing the network manually and often (incorrectly) interpreting the resulting network causally. We provide an approach to learn a Bayesian network fully from observed data, without relying on experts and show how to appropriately interpret the resulting network, both to identify how the variables (covariates and target) are interrelated and to answer probabilistic queries.3. We apply this method to the case study of a mountain pine beetle infestation and find that the trained Bayesian network has a predictive accuracy of 0.88 AUC. We classify the covariates as primary and secondary in terms of contributing to the prediction and show that the predictive accuracy does not deteriorate when the secondary covariates are missing and degrades to only 0.76 when one of the primary covariates is missing. As a complement to the previous work on constructing Bayesian networks byhand, we show that if instead, both the structure and parameters are learned only from data, we can achieve more accurate predictions as well as generate new insights about the underlying processes. K E Y W O R D Sautomatic learning, Bayesian network, invasive species, machine-learning, mountain pine beetle, pest, risk modelling, structure learning | INTRODUC TI ONPredictions are essential in aquatic and terrestrial ecology, whether the focus lies in changes in ecosystem composition, structure and richness to preserve the biodiversity and ecosystem function, or in the spatial distribution of individuals and species to inform conservation and invasive species policies. The field of predictive ecology focuses on how to make such predictions, particularly in the This is an open access article under the terms of the Creat ive Commo ns Attri bution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

We assessed the fish length – otolith length relationship (FL–OL) in Dolly Varden (Salvelinus malma malma) to verify proportional growth. A decoupling was detected during first ocean migration where fish growth was occurring at a greater rate than otolith growth. Because of this decoupling, the application of traditional back-calculation models overestimated the size-at-age in premigratory char. We developed modified back-calculation equations from existing traditional models to account for this decoupling based on discontinuous piecewise regressions. The new biological intercept breakpoint method (BI–BP) provided the most accurate representation of fish size-at-age throughout all life history stages when compared with known size-at-capture values in fish. The decoupling indicates that factors other than somatic growth are important for otolith accretion. Physiological changes during smoltification likely alter calcium uptake and thereby affect calcium deposition rates on otoliths during this short but biologically critical time period of life history. It is probable that species exhibiting similar complex ontogenetic shifts in life history will likely exhibit decoupling to some extent in the FL–OL relationship.

The rate of human-induced environmental change continues to accelerate, stimulating the need for rapid and science-based decision making. The recent availability of cyberinfrastructure, open-source data and novel techniques has increased opportunities to use ecological forecasts to predict environmental change. But to effectively inform environmental decision making, forecasts should not only be reliable, but should also be designed to address the needs of decision makers with their assumptions, uncertainties, and results clearly communicated. To help researchers better integrate forecasting into decision making, we outline ten practical guidelines to help navigate the interdisciplinary and collaborative nature of forecasting in social-ecological systems. Some guidelines focus on improving forecasting skills, including how to build better models, account for uncertainties and use technologies to improve their utility, while others are developed to facilitate the integration of forecasts with decision making, including how to form effective partnerships and how to design forecasts relevant to the specific decision being addressed. We hope these guidelines help researchers make forecasts more accurate, precise, transparent, and most pressingly, useful for informing environmental decisions.

No abstract