Virtual habitats, fossil preservation, and estimates of dinosaur biodiversity in the Cretaceous of North America

The archive of fossils and the sedimentary rocks that enclose them preserves a long record of climate change and associated fluctuations of biodiversity. However, this record is biased, and considerable efforts have focused on developing sampling methodologies and statistical treatments to mitigate for bias in a way that will allow us to understand how biodiversity has changed through time. The methods presented here attempt to account for the problem of data absence and spatial heterogeneity in fossil occurrence data. These approaches combine fossil occurrence data and palaeogeographic maps with Earth system and Ecological Niche Models (ENM). These methods are tested against three significant and contentious scientific questions regarding the evolution and extinction of non-avian dinosaurs. In the lead-up to the Cretaceous/Paleogene (K/Pg) mass extinction (66 million years ago), dinosaur diversity is argued to have been either in long-term decline or thriving until their sudden demise. With respect to North America, ENM shows a habitability decrease in areas with present-day rock outcrop from the Campanian to Maastrichtian (i.e. the last stages of the Cretaceous), which mirrors a decline in species. However, a continent-wide projection demonstrates that habitat was either stable or increasing during the Campanian-to-Maastrichtian but was not preserved. This suggests that Maastrichtian North American dinosaur diversity is underestimated, with the apparent long-term, climatically-driven decline being due to sampling bias Climate and ENM simulations focused on the K/Pg event strongly suggest that the extinction of non-avian dinosaurs was forced by asteroid impact. The experiments also show that Deccan volcanism was a mitigating event that potentially protected against an even worse extinction and supported biotic recovery. Potential drivers of the latitudinal distribution of Late Cretaceous (100.5–66 million years ago) dinosaur diversity are also evaluated. Multivariate space quantification of palaeogeographic and palaeoclimatic occupation show that sauropod hypervolumes are more constrained by temperature extremes than the other two dinosaur clades. These results suggest a different thermophysiological strategy for sauropods, with a possible explanation for the gigantothermic physiology experimented by this dinosaur group. The work presented herein highlights the potential role for multi-disciplinary workflows in understanding big macroecological questions in deep time. As with all experimental and analytical procedures it is imperative that the limitations of the methods used are tailored to the nature of the data and questions being asked.Open Acces