Search

Species interactions play a fundamental role in ecosystems. However, few ecological communities have complete data describing such interactions, which is an obstacle to understanding how ecosystems function and respond to perturbations. Because it is often impractical to collect empirical data for all interactions in a community, various methods have been developed to infer interactions. Machine learning is increasingly being used for making interaction predictions, with random forest being one of the most frequently used of these methods. However, performance of random forest in inferring predator‐prey interactions in terrestrial vertebrates and its sensitivity to training data quality remain untested. We examined predator–prey interactions in two diverse, primarily terrestrial vertebrate classes: birds and mammals. Combining data from a global interaction dataset and a specific community (Simpson Desert, Australia), we tested how well random forest predicted predator–prey interactions for mammals and birds using species' ecomorphological and phylogenetic traits. We also tested how variation in training data quality – manipulated by removing records and switching interaction records to non‐interactions – affected model performance. We found that random forest could predict predator–prey interactions for birds and mammals using ecomorphological or phylogenetic traits, correctly predicting up to 88 and 67% of interactions and non‐interactions in the global and community‐specific datasets, respectively. These predictions were accurate even when there were no records in the training data for focal species. In contrast, false non‐interactions for focal predators in training data strongly degraded model performance. Our results demonstrate that random forest can identify predator–prey interactions for birds and mammals that have few or no interaction records. Furthermore, our study provides guidance on how to prepare training data to optimise machine learning classifiers for predicting species interactions, which could help ecologists 1) address knowledge gaps and explore network‐related questions in data‐poor situations, and 2) predict interactions for range‐expanding species. [ABSTRACT FROM AUTHOR]

Extinctions stemming from environmental change often trigger trophic cascades and coextinctions. Bottom–up cascades occur when changes in the primary producers in a network elicit flow‐on effects to higher trophic levels. However, it remains unclear what determines a species' vulnerability to bottom–up cascades and whether such cascades were a large contributor to the megafauna extinctions that swept across several continents in the Late Pleistocene. The pathways to megafauna extinctions are particularly unclear for Sahul (landmass comprising Australia and New Guinea), where extinctions happened earlier than on other continents. We investigated the potential role of bottom–up trophic cascades in the megafauna extinctions in Late Pleistocene Sahul by first developing synthetic networks that varied in topology to identify how network position (trophic level, diet breadth, basal connections) influences vulnerability to bottom–up cascades. We then constructed pre‐extinction (~80 ka) network models of the ecological community of Naracoorte, south‐eastern Sahul, to assess whether the observed megafauna extinctions could be explained by bottom–up cascades. Synthetic networks showed that node vulnerability to bottom–up cascades decreased with increasing trophic level, diet breadth and basal connections. Extinct species in the Naracoorte community were more vulnerable overall to these cascades than were species that survived. The position of extinct species in the network – tending to be of low trophic level and therefore having relatively narrow diet breadths and fewer connections to plants – made them vulnerable. However, these species also tended to have few or no predators, a network‐position attribute that suggests they might have been particularly vulnerable to new predators. Together, these results suggest that trophic cascades and naivety to predators could have contributed to the megafauna extinction event in Sahul. [ABSTRACT FROM AUTHOR]