Search bioRxiv⌕ Search

Biology subjects

Cooper, R. B.

Publications and source records attributed to Cooper, R. B..

2 recordsLinked to original sources

Splendid isolation with migration: Diversity dynamics of South American mammals

Biodiversity dynamics following encounters between long-isolated faunas provide natural experiments to examine how diversity dependence and ecological interactions shape diversification patterns at macroevolutionary time scales. The Cenozoic arrival of African and North American mammals in South America likely had transformative effects on one of the worlds most unique and previously isolated continental faunas. However, the extent and persistence of these effects remain debated. Here, we assess the impact of immigrant mammal groups on native South American mammal diversity by combining an extensive fossil dataset (6,214 occurrences; 1,739 species) with deep learning methods and time-series analyses to model dynamics of mammalian biodiversity throughout the Cenozoic. The arrival of African and North American lineages increased the continents overall species richness but also contributed to a decline in native diversity. Additionally, our models suggest that the negative effect of immigrant lineages on native diversity was strongest immediately after their arrival and initial diversification, but this effect lessened over time. Overall, these findings support a scenario where immigration simultaneously enriched South American faunas and triggered a time-decaying replacement of native lineages, reshaping continental biodiversity.

evolutionary biology↗

DeepDiveR - A software for deep learning estimation of palaeodiversity from fossil occurrences

O_LIThe incompleteness of the fossil record, in particular variation in preservation and sampling through space and time, presents a barrier to estimating changes in biodiversity which standard statistical methods struggle to account for. C_LIO_LIHere we present DeepDiveR, an R package for the DeepDive program enabling estimation of biodiversity from fossil occurrence data. The method uses a simulation-trained deep neural network to generate predictions of biodiversity change through time, while accounting for temporal, spatial and taxonomic heterogeneities in preservation. C_LIO_LIDeepDiveR can be readily used to explore the extinct biodiversity of different clades. We demonstrate the pipeline to build and customise analyses, including consideration of changes in biogeography. We also further develop the model to integrate information about modern diversity in the case of extant clades and introduce a function that automatically adjusts the parameterization of the simulations to generate training data that reflect the distribution of empirical datasets. C_LIO_LITo demonstrate the software, we analyse the fossil record of the order Carnivora through the Cenozoic, finding a peak in diversity in the Late Miocene and a 37% species loss since the Pleistocene. Our implementation includes the generation summary statistics and plots that allow for an evaluation of the model performance and diversity estimations and a configuration file that captures all parameters required to guarantee the full reproducibility of the results. C_LI

paleontology↗