T150 Virtual Paleontology: Computer-Aided Analysis of Fossil Form and Function

Tuesday, 6 November 2012
8:00 AM - 12:00 PM
217A

FINITE ELEMENT ANALYSIS AND ITS APPLICATION TO PALEONTOLOGY

Paper number: T150-13
Program: Technical Sessions
Time: 11:10 AM – 11:30 AM

Authors

Abstract

From origins in industrial engineering, Finite Element Analysis (FEA) has rapidly been adopted by the life sciences. FEA is a computational technique that is able to calculate mechanical stresses and strains in complex structures: those having highly complex geometry, composite material properties, and loading from multiple directions or sources. These features make it ideal as a means of assessing the relationship between biomechanical function and skeletal form. Virtual geometries can be captured via scanning technologies, and for paleontologists, a well-made finite element model can ask questions of extinct animals that are not accessible with any other technique. Comparisons between models of extinct and extant taxa can highlight similarities and differences in biomechanical performance, and may provide glimpses of extinct behaviours. By working with computer models, evolutionarily novel structures such as ridges of thickened bone, spines, or other changes in shape may be easily added, removed or otherwise altered. This allows us to assess what, if any, contribution they make to mechanical performance and hypothesize on adaptive potential and unexplored pathways in morphological and functional evolution. Recent endeavours to combine Finite Element Analysis with statistical analysis of shape using Geometric Morphometrics (GMM) are driving forward such investigations. It is likely that many future biomechanical studies will utilize FEA, GMM, or both.

Finite Element models in paleontology offer many opportunities. However, models are simplifications of reality. Biological structures are notoriously complex, and in the production of any model such complexity must be reduced. How can we know that biomechanical FE models are good representations of reality? Furthermore, models can only be as good as the information used to create them. In paleontology, where crucial input information may be missing, how does this affect the questions that can be asked? And what therefore are the limitations of the technique? Examples from both invertebrates and vertebrates will be discussed, with several mammalian and avian examples highlighted.