171-10 Easier, Better, Faster, Stronger Lithic Identification: Combining Hyperspectral Field Imaging Analysis with Structure from Motion Photogrammetry to Quantify Entrained Lithics in Volcanic Structures Using Tsezhin Bii as a Case Study

Session: Advancing Geologic Analysis with Digital Outcrops and Close-Range Remote Sensing Data

Presenting Author:

Elisheva Sherman

Authors:

Sherman, Elisheva¹, Greenberger, Rebecca², Kerber, Laura³, Clarke, Amanda B.⁴, Arrowsmith, Ramon⁵, Hank, Jason⁶, Semken, Steve⁷, Mohr, Kyle⁸

(1) School of Earth and Space Exploration, Arizona State Univeristy, Tempe, AZ, USA, (2) California Institute of Technology, pasadena, CA, USA, (3) NASA, Jet Propultion Lab, La Canada, CA, USA, (4) School of Eaarth and Space Exploration Arizona State University, Arizona State University, Tempe, AZ, USA; INGV -Sezione di Pizsa, Pisa, Italy, (5) School of Earth and Space Exploration, Arizona State University, Tempe, AZ, USA, (6) None, White Cone, Navajo Nation, USA, (7) School of Earth and Space Exploration, Arizona State University, Tempe, AZ, USA, (8) School of Earth and Space Exploration, Arizona State University, Tempe, AZ, USA,

Abstract:

Maar-diatremes form when rising magma interacts with water, typically groundwater or ground ice, to cause phreatomagmatic eruptions. Typically, the explosive sources of these eruptions are in the subsurface and deepen over time, resulting in a conical subsurface feature (diatreme) and a subaerial excavated feature bounded by ejecta (maar). We establish a new rapid and remote method for characterizing the distribution of explosive products in maar-diatreme systems by collecting and analyzing 1) hyperspectral scans and 2) uncrewed aerial system photographs for structure-from-motion photogrammetry of exhumed maar-diatreme deposits in the Tzeshin Bii volcanic field, located on the Navajo Nation in Arizona. Tsezhin Bii was chosen as a test site due to its age, exposure, accessibility, and its abundance of documentation in geologic maps and detailed stratigraphic studies. We used Caltech’s Headwall Inc. hyperspectral imaging field spectrometer, rapidly capturing outcrop characteristics including entrained clasts of the surrounding geology (lithics). Precise identification of lithics was made possible by combining spectral analysis of the images,  knowledge of the stratigraphy and lithology of the preexisting country rock, with the high-resolution 3D models of the outcrops. Specifically, we processed images in IDL’s software ENVI, which allowed us to rapidly identify lithics at each outcrop. We also used georeferenced drone images to create 3D models of each outcrop.  The two processed data types (3D models and processed spectral images) were overlain to validate the presence of all the spectrally identified lithics and characterize their 3D spatial distribution. The field imaging spectrometer is capable of accurately identifying lithic clasts as small as 20 cm on average from approximately 60 meters away, and the upper two subsurface units, the Bidahochi and Moenave Formations, are more abundant than the deeper Chinle and Moenkopi Formations in the volcanic deposits. Additionally, the Bidahochi Formation was mixed throughout all outcrops, including into the deep root zone of the diatreme. While our lithic identification is thus far consistent with previous detailed field work, we also note that we were able to identify more lithics over a broader area using this new method. We aim to establish detailed protocols for completing rapid field mapping in locations that are remote or less secure. The more detailed tracking of lithics from their source to their current position can refine the eruption dynamics model.

Geological Society of America Abstracts with Program. Vol. 57, No. 6, 2025

doi: 10.1130/abs/2025AM-10350

© Copyright 2025 The Geological Society of America (GSA), all rights reserved.