254-8 Deep Learning Reveals Complex Sediment Penetrative Burrows from the Terminal Ediacaran of Namibia
Session: Recent Advances in Fossil Imaging
Presenting Author:
Katherine TurkAuthors:
Turk, Katherine A.1, Darroch, Simon2, Edie, Stewart M.3(1) Department of Paleobiology, National Museum of Natural History, Washington, D.C., USA, (2) Senckenberg Museum of Natural History, Frankfurt, Germany, (3) Department of Paleobiology, National Museum of Natural History, Washington, D.C., USA,
Abstract:
Sediment penetrative burrowing is a critical process that exerts significant control on seafloor structure, chemistry, and ultimately, habitability. This behavioral innovation has historically been associated with the early Cambrian; however, in recent years late Ediacaran sites have yielded several notable examples of these behaviors, thus suggesting the evolutionary roots of the modern, animal dominated benthos extend further back than previously thought. However, assessment of these putative early trace fossils is often limited by their ephemerality, preservation, and relative lack of diagnostic form. As such, deep learning approaches to image analysis represent a promising avenue for exploration of these structures -- and behavioral innovations -- in 3D.
Here we provide evidence for late Ediacaran (c. 539.41 Ma) sediment penetrative burrow-like structures from the basal Spitskop Member of the Nama Group, Namibia, segmented from CT scans with a 2D U-Net model. These structures, found in association with serially probing worm traces, bear a variety of forms, including spirals, with many at least 4 cm deep. As such, these structures potentially represent among both the earliest and deepest examples of sediment penetrative burrowing, and as such bear significant import for our understanding of the timing of these critical ecosystem engineering behaviors prior to the Cambrian boundary.
Geological Society of America Abstracts with Program. Vol. 57, No. 6, 2025
doi: 10.1130/abs/2025AM-11189
© Copyright 2025 The Geological Society of America (GSA), all rights reserved.
Deep Learning Reveals Complex Sediment Penetrative Burrows from the Terminal Ediacaran of Namibia
Category
Topical Sessions
Description
Session Format: Oral
Presentation Date: 10/22/2025
Presentation Start Time: 10:15 AM
Presentation Room: HBGCC, 304B
Back to Session