Primary Supervisor: Margarete Jadamec, PhD

Secondary Supervisor: Matt Knepley, PhD

The Aleutian-Alaska subduction zone forms the northern boundary of the Pacific Ring of Fire. The eastern region of the subduction zone is characterized by flat slab subduction, active plateau collision-subduction, significant faulting along the megathrust (the 1964 Great Alaska earthquake, Mw 9.2), and large-scale intra-continental faulting along the Denali fault (Mw 7.9 earthquake in 2002). However, untangling the relative contributions of these tectonic components on the stress state in south-central Alaska, the state’s most populous region, remains a challenge due to the three-dimensional (3D) complexity of the tectonic framework. The project examines the results of large-scale 3D numerical simulations to identify the relative contribution of flat slab subduction, plateau subduction-collision, coupling along the plate interface, and motion along Denali fault shear zone on the stress state of south-central Alaska. Students will gain experience programming in MATLAB, navigating the UNIX/Linux interface, and working with output from data-driven simulations run on high-performance computing systems.

Primary Supervisor: Stuart Evans, PhD

Secondary Supervisor: Elizabeth Thomas, PhD

Lake effect snow is the major climate hazard of Western New York (WNY). Winter of 2022 was an exceptional demonstration of the scale of human impacts and need for planning around this annual hazard. Many questions yet remain about predicting the variability of snowfall and the contribution of moisture from the lakes.  Students on this project will relate snowfall records in WNY, especially extreme snowfall, to a variety of other geophysical and environmental data using data processing languages (Matlab, R, or Python).

Primary Supervisor: Stuart Evans, PhD

Secondary Supervisor: Andrew Crooks, PhD

Dust storms have major impacts on human health and infrastructure, but are difficult to detect by traditional means of environmental monitoring.  This project will mine news media archives to produce GIS maps of newsworthy dust storms around the world, determine how frequently cities and regions are struck by dust storms, and use AI to extract additional details from this unconventional dataset.

Primary Supervisor: Austin Angulo, PhD

Secondary Supervisors: Irinia Bnedyk, PhD, Andrew Crooks, PhD

This research project investigates the challenges and experiences of cyclists navigating the University at Buffalo's North and South campuses during the harsh winter months. Given Buffalo's significant snowfall, the study will explore how factors like snow and ice accumulation, dedicated bike lane maintenance, and campus infrastructure affect the safety and accessibility of cycling. The project aims to collect both quantitative data, such as the frequency of winter cycling and preferred routes, and qualitative insights through interviews with cyclists about their perceptions of risk, strategies for managing winter conditions, and suggestions for improving the cycling environment. The findings will provide valuable information for campus planners to enhance winter cycling infrastructure and promote a year-round, sustainable transportation culture at the university.

Primary Supervisor: Austin Angulo, PhD

Secondary Supervisors: Irinia Bnedyk, PhD, Andrew Crooks, PhD

This project examines the experiences of cyclists embarking on multi-day trips along the Erie Canalway Trail, a key section of the Empire State Trail, starting from Buffalo. The research will delve into the logistical, physical, and psychological factors that influence these long-distance journeys. We'll explore how cyclists plan their routes, manage equipment and supplies, and interact with the communities and historical sites along the trail. The study aims to gather insights on the challenges of multi-day cycling, including the physical demands, safety concerns,  the solitude or camaraderie of the trail, and the unexpected discoveries along the way. By documenting these journeys, the project will contribute to a deeper understanding of recreational cycling tourism and help trail managers and local businesses better support long-distance cyclists.

Primary Supervisor: Kristin Poinar, PhD

Secondary Supervisor: 

Use an existing numeric model constrained by observations to determine how deep crevasses go on a Greenland glacier.

Primary Supervisor: Eric Sandvol, PhD

Secondary Supervisor: Grigioris Lavrentiadis, PhD

 Probabilistic seismic hazard analysis (PSHA) relies on accurate ground-motion models to quantify site-specific ground-motion distribution to guide the design of resilient infrastructure. Non-ergodic ground-motion models (NGMMs) have demonstrated clear advantages over traditional ergodic approaches by providing more realistic site-specific predictions. However, a central assumption of this methodology remains largely untested, that the more frequent, smaller-magnitude earthquakes contain information that can improve predictions of ground motions from larger, rarer events. To investigate this, we propose to use seismic network recordings from the aftershocks of the 2023 Kahramanmaraş earthquake sequence. Our study will systematically compare ground-motion spectra from both strong-motion and weak-motion data, with the goal of characterizing the nonlinear behavior of ground motions across small and large strain levels. This work represents an initial step toward bridging seismological and engineering perspectives in forecasting ground motions, ultimately enhancing the robustness of PSHA and informing resilient infrastructure design.

Primary Supervisor: Eric Sandvol, PhD

Secondary Supervisor: Grigioris Lavrentiadis, PhD

The Northeastern United States has historically received less attention in characterizing shear-wave velocity structure and site amplification compared to other regions of the country, largely due to its relatively lower seismic demand. However, the presence of critical infrastructure, combined with the increased hazard estimates in the most recent USGS updates, underscores a pressing need for improved regional characterization. This project will address that need by deploying seismic nodal arrays to collect ambient ground-motion data, which will be used to evaluate shear-wave velocity profiles and site amplification effects. The study will integrate both experimental fieldwork and computational modeling to develop region-specific site-amplification models. The outcomes will contribute to more accurate seismic hazard assessments and provide valuable input for the design and resilience of critical infrastructure in the region.

Primary Supervisor: Xudong Fan, PhD

Secondary Supervisor: Kang Sun, PhD and Yifan Cheng, PhD

Coastal flooding in shoreline and Great Lakes regions arises from the compound effects of precipitation, wind, seiche oscillations, and land–surface interactions. Conventional models such as ADCIRC, Delft3D, or HEC-RAS demand high-resolution bathymetry, boundary forcing, and long runtimes, limiting rapid prediction and scalability. This project aims to integrate NASA Earth-observing satellites—GPM (precipitation), CYGNSS (wind), and SMAP (soil moisture)—with flooding simulation models to enable AI-driven, physics-informed flooding prediction. By fusing remote-sensing data with physics-informed machine learning (PIML) surrogates of coastal models, this project aims to rapidly generate high-resolution inundation maps and dynamic flood forecasts under varying climate conditions for coastal regions. The outcome of this project is expected to provide real-time coastal flood warnings, enhance situational awareness, and support resilient infrastructure and emergency response in climate-vulnerable coastal communities.

Primary Supervisor: Xiao Lou, PhD

Secondary Supervisor:  Naveen Senthil and Sophie Nowicki, PhD

The most uncertain component of sea level change comes from the Greenland and Antarctic ice sheets. Here we explore the spatial feature of Antarctica ice evolution in 2100 and 2300 using principal component analysis applied to the state-of-the-art ice sheet model ensemble from the Ice Sheet Model intercomparison Project for CMIP6 (ISMIP6). The outcome of this project will be insight into the differences and assumptions made in the various experiments and model simulations to help refine our understanding of the future evolution that the Antarctic ice sheet may face.

Primary Supervisor: Scott Mackay, PhD

Secondary Supervisor:  Stuart Evans, PhD

Forest die-off has been accelerating due to warm droughts coupled with increased biotic epidemics and wildfire powered by fuel build-up. Forests are essential to environmental and societal health, and so an improved ability to predict forest responses to rapidly changing climate will help us develop solutions for the future. Students on this project will use plant ecophysiological data along with analytical and predictive tools to examine and explain patterns of forest die-off. They will learn how to obtain data from global vegetation databases, parameterize and run predictive models of woody plant responses to abiotic and biotic stressors, and perform statistical analyses to assess predictive ability of the models.