MATERIALS AND METHODS

A RIEGL LMS-Z420i 3D terrestrial Light Detection and Range (LiDAR) scanner was used to digitize the University of Wyoming (UW) Geological Museum's mounted cast of MOR 693 (Figure 1.1, 1.2). The CAD package Maya (www.autodesk.com/maya) was used to construct body outlines around the digital skeletal model, following the approach of Bates et al. (2009; figures 1.3, 1.4). The skull of MOR 693 was mediolaterally distorted during post-mortem burial and so was re-inflated by 20% prior to head volume reconstruction, based on observations of published descriptions of A. fragilis (Madsen 1976). Each reconstructed body segment was given a density of 1000 kg m^-3, in accordance with previous studies (Alexander 1985, 1989; Henderson 1999; Hutchinson et al. 2007).

After producing a 'best estimate model' (see discussion below) a sensitivity analysis was conducted to investigate the effects of initial assumptions and estimate a realistic range of mass set values for MOR 693. Mass properties were calculated for a single gracile (minus 7.5%) and two larger models (+ 7.5% and +15%), by changing the diameter of the NURBs circles that defined the neck, thoracic, sacral, tail and hind limb segments (thigh, shank and metatarsal). Bates et al. (2009) found that the + 15% models far exceeded the likely maximum body segment volumes for the five animals studied. In an attempt to better constrain the likely maximum mass set values for MOR 693 an additional + 11.25% model was constructed. To investigate uncertainties regarding the relative proportions of body segments, a series of trunk and leg segments from the respective models were interchanged. In addition to segment volumes, the effects of having larger and smaller zero density respiratory structures in our thoracic and neck segments were also tested.

Even near-complete specimens suffer from uncertainties about precise articulation of the skeleton, yet no physical or digital mass reconstruction to-date has tested for the effects of these unknowns on mass set predictions (Bates et al. 2009). As a first step towards quantifying uncertainties in mass predictions related to the mounting of dinosaur skeletons additional models were produced in which the articulation of trunk elements were varied. Specifically, the effects of two uncertainties in dinosaur skeletal reconstructions are tested; the medio-lateral angulation or flare of the ribcage and relative spacing of individual vertebrae (i.e., inter-vertebral disc length). First, the best estimate thoracic volume was adjusted after increasing and decreasing the orientation or flare of the rib cage +/- 10 degrees. Second, the best estimate reconstruction was modified by increasing the separation between vertebrae and their associated NURBs outline by 0.005 m to mimic an enlarged inter-vertebral disc cavity. The original spacing of vertebrae from the dorsal and caudal series in the casted skeleton can be found in the Appendix (Table 26).