Future Palaeontologists Will Detect Current Mammal Latitudinal Biodiversity Gradient

Résumé fourni par la source

ABSTRACT Aim Fossil data provide crucial insights into past biogeographic and macroecological patterns. However, geological, biological, and sampling biases can potentially compromise genuine biodiversity inferences. Here, we tested whether fossil biases may hinder the accurate retrieval of the Latitudinal Biodiversity Gradient (LBG). Location Global. Time Period Contemporary. Major Taxa Studied Mammals. Methods We implemented a filtering process to current mammal distribution maps, simulating one geological, two biological, and three sampling sources of bias. Namely, distribution maps were downgraded to regions with sediments, species preservation was modulated by their range size and body size, and sampling was applied to locations with a fossil record. We applied the filters sequentially to mimic a process of progressive fossilisation, considering three preservation rates and removing up to 98.8% of the original species. We also applied filters independently to assess their individual effect. Lastly, we quantified the richness loss, the change in the slope between latitude and richness, and the change in richness maxima throughout the filters. Results Results indicate that the applied filters collectively and distinctly influence the detection and robustness of the LBG signal. However, although the slope of the richness gradient diminishes progressively (especially for filters affecting species by their body size or taxonomic group), a LBG signal is detected across all the filters. Equally, despite the critical species loss, richness maxima remain around the equator. Main Conclusions We demonstrated that strongly incomplete or biased samples can still recover accurate large‐scale biogeographic patterns such as the LBG. Our results show an optimistic scenario in which, although the LBG intensity is sensitive to the uneven loss of information in biodiversity data, a detectable signal can be retrieved for all scenarios.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.