David Barrick

Date of Award

Spring 5-6-2022

Degree Type

Non-Thesis Project

Degree Name

MS Geosciences/Geochemistry


Geological Engineering

Committee Chair

Mary M. MacLaughlin

First Advisor

Curtis A. Link

Second Advisor

Mohammad Sadeghi


This paper summarizes the results of a suite of tests performed in conjunction with a site investigation for the replacement of six timber bridges in Central Montana and four investigative geotechnical boreholes from a nearby ranch. The sites at the six timber bridges consist of Cretaceous sedimentary rocks (shale with stratified sandstone), overlain by thin layers of residual clays or Quaternary alluvium. The geology at the nearby ranch consists of either Quaternary vuggy travertine or Quaternary alluvium/colluvium deposits overlying residual shale of the Mississippian Heath Formation. In total, 49 samples were collected ranging from three Quaternary travertine HQ core samples, 12 Shelby tube samples of overlying clay, and 34 HQ core samples of Cretaceous sedimentary rocks. Engineering characterization of these samples was done via basic laboratory tests, including standard unconfined compression (UCS) tests performed using Montana Tech’s TerraTek servo-hydraulic load frame. These tests yielded values of static modulus, moisture content, unit weight, and peak compressive strength. Data from the specimens are correlated to tomograms from field geophysical investigations using four geophysical methods: 2D compressional-wave refraction tomography (SRT), 2D shear-wave multi-channel analysis of surface waves (MASW), 2D electrical resistivity tomography, and 2D induced polarization tomography. Geophysical values were obtained from the final 2D images and compared with geotechnical borehole data at the same depths. These geotechnical and geophysical values are then analyzed using a Feedforward Neural Network (FNN). The goal of FNN analysis is to determine whether a relationship can be mapped between the geophysical and geotechnical values of these sedimentary rocks. Establishing such a relationship would allow future investigation via nondestructive experimental techniques and provide confidence in using geophysical site investigations where geotechnical drilling and coring cannot be done. The results from the FNN analysis suggest a high confidence of correlating desired common geotechnical parameters from the independent geophysical variables for sedimentary rocks in Central Montana.