Search bioRxiv⌕ Search

Biology subjects

Llewelyn, J. S.

Publications and source records attributed to Llewelyn, J. S..

2 recordsLinked to original sources

Trait-space disparity in fish communities spanning 380 million years from the Late Devonian to present

The diversity and distribution of species traits in an ecological community determine how it functions. While modern fish communities conse rve trait space across similar habitats, little is known about trait-space variation through deep time or across different habitats. We examined how fish trait diversity varies through space and time by comparing three Late Devonian fish communities -- a tropical reef (Gogo, Australia), a tropical estuary (Miguasha, Canada), and a temperate freshwater system (Canowindra, Australia) -- with six modern communities from diverse habitats. Trait-space metrics reflecting within-community diversity (functional richness) and species similarity (functional nearest-neighbour distance) indicated Late Devonian communities had scores similar to modern communities. However, they were less functionally rich than their closest modern analogues, and their species tended to be more functionally distinct from one another. Metrics describing location in trait space (centroid distances and hypervolume overlap) showed modern communities were similar to each other, Gogo and Miguasha were similar but distinct from modern communities, and Canowindra was distinct from all others. This pattern suggests period-associated differentiation and substantial heterogeneity among some Late Devonian communities. In addition to temporal changes, we found consistent differences associated with habitat type and climate zone. Reef and tropical communities were the most functionally rich, whereas functional nearest-neighbour scores were highest in estuarine and temperate communities. These results indicate fish community trait space varies with time, habitat and climate, suggesting (i) lability in fish trait space and (ii) that evolutionary history, environmental filtering, and stochasticity influence community assembly.

evolutionary biology↗

Predicting predator-prey interactions in terrestrial endotherms using random forest

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 (i) address knowledge gaps and explore network-related questions in data-poor situations, and (ii) predict interactions for range-expanding species.

ecology↗