Evidence for taphonomic size bias in the Dinosaur Park Formation (Campanian, Alberta), a model Mesozoic terrestrial alluvial-paralic system

A study of the distribution of dinosaurian body masses in the Dinosaur Park Formation (DPF; Campanian; southern Alberta), reveals a prominent negative skew; a pattern distinct from those of modern terrestrial faunas. We find a direct and robust correlation between taxon size (estimated live body mass) and known completeness. There is a clear dichotomy between large and small-bodied taxa at around 60 kg, in which taxa less than 60 kg are significantly less complete (mean completeness = 7.6%) than those with an estimated mass of 60 kg or greater (mean = 78.2%). Along with completeness, there is also a strong association of body size and taphonomic mode, with small taxa known largely from isolated and occasionally associated remains, and large taxa known from articulated skeletons. In addition, there is a significant correlation between taxon body mass and both date of discovery and of description, with taxa <60 kg taking an average of 65.9 and 75.6 years to discover and describe, respectively, compared to 33.6 and 34.1 years for taxa >60 kg. The rates of both cumulative discovery and description for large taxa are best described by a logarithmic curve nearing an asymptote, whereas small taxa show either a linear or power increase through time. This suggests that our current knowledge of the large-bodied dinosaur assemblage is reasonably representative of the true biological fauna with few discoveries likely to be made in the future. However, small taxa are greatly underestimated in both their diversity and abundance, with many more potential discoveries to be made. Given that (1) the sedimentary deposits and fossil assemblages in the DPF together represent one of the best studied examples of a Mesozoic alluvial-paralic (terrestrial) 'palaeoecosystem,' and (2) similar patterns have been suggested (but not documented) for other Mesozoic terrestrial ecosystems in the Western Interior of North America, we suggest that this pattern of size bias may typify vertebrate fossil assemblages in terrestrial Mesozoic systems. If so, such biases must be considered before patterns of diversity in dinosaur communities through time can be considered accurate, or used to compare and interpret Mesozoic palaeoecosystems. (C) 2012 Elsevier B.V. All rights reserved.