I am always eager to collaborate with motivated and enthusiastic students interested in contributing to my research projects. If you’d like to explore the types of problems I focus on, please visit my research page or review the examples of past and ongoing ASU student projects outlined below.

While I value students with a strong mathematical background, programming experience (e.g., GES 3455), and familiarity with Linux/Unix operating systems, these skills are not always prerequisites. What matters most is a willingness to learn and a passion for using computers to visualize and solve Earth science problems through the principles of physics, mathematics, and computer science.

If you’re curious about the nature of student projects or have questions about getting involved, feel free to stop by my office or send me an email at marshallst@appstate.edu.

Note: While I have had the pleasure of working with graduate students at other institutions, this page focuses exclusively on ASU student projects.

Rachel worked with Dr. Sarah Evans and me at a permafrost site in the North Slope region of Alaska. She collected two seasons of in situ measurements and Ground-Penetrating Radar (GPR) data across a water track and found several enigmatic features in the data. To explain the features, she also created computational models to demonstrate that some of the complex reflection pattern in GPR data we found are likely due to buried triangular ice wedges. Our work was published in the Journal of Geophysical research in 2024.

Lanie completed a project that was very similar to what an environmental consulting company would do. She used Ground-Penetrating Radar (GPR) and direct current electrical resistivity imaging along 7 transects taken around the old Broyhill Hotel site on the ASU campus. Her data showed the depth to bedrock and variations in the saprolitic bedrock properties at the site.

Shams used direct current (DC) electrical resistivity methods to detect and characterize caves in the Water Sinks, VA area. Shams tested the effects of varying smoothness and noise thresholds using a 2D inversion modeling software. His data not only confirmed known void spaces near major caves in the area, it identified previously unknown void spaces which are likely to effect grouwater flow paths.

Max wrote a MATLAB algorithm that computes strain rate inversions of GNSS velocities. He applied the algorithm to GNSS data in southern California and found significant non-tectonic sources of strain including urban or agricultural water use. These anthropogenic motions produce strain that are most commonly dilatational in nature and contaminate several localized regions including the Los Angeles and Ventura basins and the San Joaquin Valley. In an attempt to minimize non-tectonic motions, utilized the gpsgridder function, which interpolates the velocities on a uniformly spaced grid using the Green’s functions for a thin elastic plate. Max found that gpsgridder appears to reduce the resultant anthropogenic strain rates and results in spatially smoother strain rate maps. This suggests that anthropogenic motions may contain a significant component of non-elastic motion.

Savannah created and implemented a new method for driving deformation in 3D mechanical models in southern California. The method drives deformation by the total Pacific-North American plate motion as well as slip on the San Andreas, San Jacinto, and Garlock faults. Based these boundary conditions, she solved for the full 3-dimensional distribution of slip on 83 faults throughout southern California. Her work also suggests that the San Gabriel, Pine Mountain, and Big Pine Faults may be currently inactive. Her work was published in the Seismological Society of America in 2022.

Jayson created 3D mechanical models of active faulting in the Imperial Valley, CA based on the Southern California Earthquake Center's Community Fault Model. He tested several different geometries for the Brawley Seismic Zone faults and compared results to both long term slip rates and current-day interseismic GNSS and InSAR data.

Taylor used Ground-Penetrating Radar (GPR) and direct current electrical resistivity surveys to constrain a three-dimensional model of a floodplain along the New River in Boone, NC. By iteratively adjusting GPR velocities and comparing identical features to those imaged with resistivity data, Taylor was able to effectively calibrate the GPR velocity to achieve accurate depth scales. The resultant data was used to constrain a three-dimensional model of the subsurface materials.

Jacob and collaborator, Betsy Madden, created 3D Boundary Element Method mechanical models of active faulting in the Imperial Valley, CA. We used these models to predict the long-term slip rates of the regional faults, many of which are otherwise unconstrained. Because some of the fault geometries are poorly constrained, we tested a range of different geologically-plausible fault configurations and found that only one fault configuration can produce slip rates that match existing geologic slip rate data. The results of Jacob's work were published in Bulletin of the Seismological Society of America in 2019.

Hannah processed GNSS timeseries for 347 sites throughout southern California and compared results to rainfall data from numerous southern California monitoring stations. She found that seasonal amplitudes of surface motion observed in GNSS Data do not significantly vary spatially, even between significantly differing climates within southern California. She was able to identify significant non-tectonic motions in 22/347 continuous GNSS stations throughout southern California.

Adam worked to characterize the temporal variations in subsurface electrical resistivity in a local floodplain aquifer system. To do this, Adam ran numerous direct current electrical resistivity surveys at the same location throughout a one-year period. Adam inverted the survey data and calculated statistics of the results which he compared to in-situ measurements in a nearby borehole. He showed that median resistivity values follow an inverse relationship with temperature, and that days of high water table produce anomalous results.

Hugh worked with me to create computational models of fault slip rates throughout the greater Los Angeles region, CA. We had never simulated this many faults (64) before and so his work pushed forward our modeling methodology. Hugh showed that subsurface fault intersections exert a primary control on the distribution of slip. Hugh's work was published in the Seismological Society of America in 2022.

Processing InSAR data requires numerous parameters that must be quantified, but are often non-physical. Jay's project involved testing all of these parameters to better understand the effects on the resultant data. He then produced a datset of surface motions from the InSAR data and identified numerous locations experiencing anthropogenic deformation in southern California. Jay's work was presented at the fall meeting of the American Geophysical Union in 2014.

Julia worked with me along with collaborators Gareth Funning and Chad Severson at the University of California Riverside to create mechanical models of the stress changes and the potential for triggered slip on the San Andreas due to the Northridge earthquake. Julia's work also involved writing an analytical dislocation modeling code based on Okada (1984) using the prgramming language, MATLAB. Julia presented her results at the American Geophysical Union fall meeting in December of 2011 and work won the outstanding visualization and computation award at the ASU student research symposium in April 2012.

Jesse worked with me along with Ellen Cowan to image the subsurface geologic deposits of a set of fluvial terraces along the south fork of the New River in northwest North Carolina. Jesse used a combination of Ground-Penetrating Radar (GPR) and Direct Current (DC) electrical resistivity surveying to provide a detailed picture of the near surface geology of the terraces. This work was presented at the Southeast Section meeting of the Geological Society of America in 2012.

Bevin used near-surface geophysical data to characterize the subsurface geology of the floodplain along the south fork of the New River in Boone, NC. She collected, processed, and integrated data from Ground Penetrating Radar (GPR) and Direct Current (DC) electrical resistivity to image the subsurface in three-dimensions. Bevin's work also involved assessing the effectiveness of each technique based on a comparison of the geophysical data to borehole data. Bevin's results were presented at the 2010 annual meeting of the American Geophysical Union in San Francisco, CA. Her work has paved the way for what is now an annual class project in my GES3160 "Intrduction to Geophysics" course.

Anna Morris, a graduate student in the department of physics, modeled the slip behavior and seismic potential of wavy fault surfaces. Faults are inherently non-planar geologic structures; however, most seismic hazard analyses utilize highly simplified fault geometries. To determine the effects of fault geometry on slip behavior and seismic potential Anna and I created a suite of numerical models of sinusoidally wavy fault surfaces and found some interesting results. The results of this work were published in the Journal of Geophysical Research in March of 2012.