The role of sampling, environment and biotic interactions on the distribution and diversification of Mesozoic dinosaurs

Dinosaurs dominated terrestrial ecosystems for over 140 million years, making them one of the most important clades for understanding large-scale patterns in the evolution of terrestrial diversity. Although they are among the most intensively studied vertebrate groups, their fossil record is unevenly sampled through time and space, distorting our understanding of both the group’s origin and their subsequent diversification throughout the Mesozoic. Correcting for such biases is therefore essential for extracting genuine macroevolutionary signals and for evaluating how ecological and environmental processes shaped the long-term diversity and distribution of terrestrial vertebrates. In this thesis, I combine a suite of quantitative methods to reassess the origin and diversification of the dinosaurs whilst explicitly accounting for spatiotemporal sampling biases. First, I use sampling-corrected biogeographic models to test competing hypotheses of dinosaur origins within the broader archosaur radiation. These analyses support a low-palaeolatitude Gondwanan origin for dinosaurs, a result obscured in previous work by uneven preservation in southern tropical regions. I then investigate diversity dynamics by combining occurrence-based subsampling approaches with Bayesian modelling of the fossil record, applied to both global and regional datasets, alongside phylogeny-based diversification analyses. These results demonstrate that dinosaur diversity did not arise from a single explosive radiation but instead unfolded through multiple pulses of diversification across the Mesozoic, plausibly linked to major environmental perturbations. In some cases, however, Lagerstätten continue to exert a strong influence on perceived patterns, highlighting how the spatial as well as temporal unevenness of the fossil record can disproportionately shape global diversity trajectories. The results of this thesis provide new insights into the origin and diversification of dinosaurs, demonstrating that integrating complementary methods which explicitly account for spatiotemporal sampling biases can overcome the limitations of an uneven fossil record and clarify how ecological and environmental processes have shaped biodiversity through time and space.