Dinosaur biogeographic structure and Mesozoic continental fragmentation: a network-based approach

Aim: To reconstruct dinosaur macro-biogeographical patterns through the Mesozoic Era using a network-based approach. We test how continental fragmentation affected dinosaur macro-biogeographical structure and evolutionary rates. Location: A global occurrence database of dinosaur families from the Late Triassic to the end-Cretaceous was used for this study. Methods: Biogeographical and geographical network models were constructed. Continental landmasses were linked by direct continental contact and sea level (SL)-conditioned connections in geographical networks, and by shared dinosaur families in biogeographical networks. Biogeographical networks were run with raw, novel and first-step connections for all dinosaur, ornithischian, theropod, and sauropodomorph taxa. Results: Geographical connectedness declines through time, from peak aggregation in the Triassic–Jurassic to complete separation in the latest Cretaceous. Biogeographical connectedness shows no common trend in the raw and novel connection network models, but decreases through time while showing some correlation with continental fragmentation in most of the first-step network models. Despite continental isolation and high SLs, intercontinental faunal exchange continued right up to the end of the Cretaceous. Continental fragmentation and dinosaurian macro-biogeographical structure do not share a common pattern with dinosaurian evolutionary rates, although there is evidence that increased continental isolation resulted in increased origination rates in some dinosaurian lineages. Spatiotemporal sampling biases and early Mesozoic establishment of family-level distribution patterns are important drivers of apparent dinosaur macro-biogeographical structure. Main conclusions: There is some evidence to suggest that dinosaur macro-biogeographical structure was influenced by continental fragmentation, although intercontinental exchange of dinosaur faunas appears to have continued up to the end of the Cretaceous. Macro-biogeographical patterns are obscured by uneven geographical sampling through time and a residual earlier Mesozoic distribution which is sustained up to the end of the Cretaceous.