Analyzing taphonomic deformation of ankylosaur skulls using retrodeformation and finite element analysis

Abstract

Taphonomic deformation can make the interpretation of vertebrate fossil morphology difficult. The effects of taphonomic deformation are investigated in two ankylosaurid dinosaur taxa, Euoplocephalus tutus (to investigate effects on our understanding of intraspecific variation) and Minotaurasaurus ramachandrani (to investigate the validity of this genus). The ratio of orbit maximum rostrocaudal length to perpendicular height is used as a strain ellipse, which can be used to determine if ankylosaur skull fossils have been dorsoventrally compacted during fossilization and diagenesis. The software program Geomagic is used to retrodeform three-dimensional (3D) digital models of the ankylosaur skulls. The effects of sediment compaction are modeled using finite element analysis, and the resulting strain distributions are compared with the retrodeformed models as a test of the retrodeformation method. Taphonomic deformation can account for a large amount of intraspecific variation in Euoplocephalus, but finite element analysis and retrodeformation of Minotaurasaurus shows that many of its diagnostic features are unlikely to result from deformation.

Figure 1. Comparison of AMNH 5405 (Euoplocephalus) and INBR 21004 (Minotaurasaurus) in ventral view.

Specimens scaled to same premaxilla-occipital condyle length. Abbreviations: bs – basisphenoid, ic – internal choana, nc – nuchal crest, o – orbit, oc – occipital condyle, pmx – premaxilla, poc – paroccipital process, pt – pterygoid, q – quadrate, qjh – quadratojugal horn, sh – squamosal horn, tr – tooth row, v – vomer.

Figure 2. Measuring orbit shape, and deforming digital models in Geomagic.

A) Two dimensions were measured for each orbit, the maximum rostrocaudal length, and the perpendicular height, shown here on TMP 1999.58.79, Chelydra serpentina. B) To retrodeform digital skull models in Geomagic, the “Deform Region” tool is selected and placed at the midline of the skull, between the orbits. C) The arrow is adjusted into the desired position, in this case, pointing dorsally. D) The tool is then expanded to encompass the entire skull.

Figure 3. Results of orbit shape measurements for extant taxa.

The mean ratio for each taxon is represented by the black circle, and the standard deviation by the vertical line. The blue horizontal line shows the mean ratio for all taxa except crocodilians and lizards, and the light blue box represents the standard deviation. The mean orbit ratio is 1.14±0.14 (n  = 96).

Figure 4. Results of orbit shape measurements for ankylosaurs.

An R or L after the specimen number denotes the right or left orbit, respectively. The light blue box represents the mean orbit ratio ± one standard deviation for extant taxa (1.14±0.14).

Figure 5. Results of deformation and retrodeformation of models using Geomagic.

The top half of the image shows AMNH 5405 with (from left to right) no compression, 5 cm compression, and 8 cm compression; the rightmost column shows the original UALVP 31 skull for comparison. The bottom half of the image shows INBR 21004 with (from left to right) 8 cm retrodeformation, 5 cm retrodeformation, and no retrodeformation.

Figure 6. Results of the finite element analyses simulating taphonomic deformation in Euoplocephalus.

AMNH 5405 in oblique rostrolateral view (left column) and ventral view (right column).

Figure 7. Results of the finite element analyses simulating taphonomic deformation in Minotaurasaurus.

INBR 21004 in oblique rostrolateral view (left column) and ventral view (right column).