GSA Connects 2022 meeting in Denver, Colorado

Paper No. 273-6
Presentation Time: 2:00 PM-6:00 PM

A SURFACE SCANNING METHODOLOGY FOR ACCURATE, HIGH-RESOLUTION DIGITIZATION OF FOSSILS


BUTLER, Garrett1, HEBERER, Mikelia1, FERRILL, Thomas James2, PETERMAN, David1 and RITTERBUSH, Kathleen1, (1)Department of Geology and Geophysics, University of Utah, 115 S 1460 E, Salt Lake City, UT 84112, (2)J. Willard Marriott Library 295 S, University of Utah, 295 S 1500 E, Salt Lake City, UT 84112

Paleontological research often requires observation of small and detailed material with a careful eye. Precise and accurate measure of subtle features – from grooves on ancient shark teeth to ornament on mollusc conchs – is vital for systematics, descriptions, morphometrics and biomechanics. We present a workflow to produce and analyze three dimensional models of fossil specimens.

As a test case, we image ribbing ornament on fossil ammonoid conchs, features often ascribed ecological relevance (i.e., predation deterrence; hydrodynamic benefits). Conventional photography cannot always capture ornamentation on an ammonoid fossil. We chose 20+ scaphitid ammonite fossils from the Cretaceous Western Interior Seaway as targets for model construction and analysis. We target sufficient ribbing along the ventral shoulder to examine how an individual specimen’s rib wavelength and amplitude vary through whorl expansion.

The workflow has four parts. First, we produce 3D models with a structured light scanner mounted on a swing boom above a fossil resting on a turntable. The Artec Spider scanner uses a narrow band of blue light and a camera array to determine distances between features of an object, making a reference frame of point data. With a depth of field of 170-350 mm, a field of view 140x180mm, and produces 4-8 frames per second with 100-micron (.01mm) resolution. We then use the Artec Studio Pro software to align point clouds and filter scan data to then generate a 3D mesh (faces that span between vertices).

Second, we isolate portions of the digitized shell for analysis. We use the open source software Blender to isolate “slices” of the shell: in this case, spans of ribbing expressed across a specimen’s ventral shoulder. Third, we trace 2D images of our “slices” in ImageJ to export the shape data as xy coordinates. Fourth, we perform waveform analysis of the coordinate data (wavelength, amplitude, etc.) in R.

The 3D models allow easier isolation of nuanced shape profiles than still photography, and a single well-refined scan can serve many purposes. The Modestly-sized files are easily opened on any computer (20-200 Mb .stl files), more accessible than CT data. Future uses of these digital models include computational fluid dynamics simulations; hydrostatic calculations; and 3D printed classroom teaching specimens.