Six undergraduate students in the School of Engineering have been spending their summer at the University of Vermont collaborating with faculty on a variety of research projects. The support for this special program comes from two UVM alumni who cherish their experiences at UVM and who wanted to promote similar learning opportunities for current students.
Richard Barrett, ’66 UVM alumnus, founded the Barrett Foundation as a family nonprofit because his own career was boosted by his early undergraduate internship experience. His vision for undergraduate student research created this program, which is in its eighth year. Additionally, Ronald Liston’49 UVM alumnus, created another pair of scholarships due to his love for learning. His successful 57 year military and civilian engineering career fueled his desire to contribute to the existing Barrett Research platform, so he augmented the program by empowered members of the UVM Chapter of the Society of Women Engineers (SWE) to participate.
“These undergraduate engineering students have the unique opportunity to pursue independent research and work with leading faculty scholars in their fields,” says Dr. Donna Rizzo, faculty advisor and P.I. for the Barrett Foundation scholarship grant.
Overviews of the 2013 student research projects are in alphabetical order below:
LiDAR used to determine Sediment Loading from Streambank Erosion
Hanna Anderson (Advisors: Drs. Mandar Dewoolkar and Donna M. Rizzo)
In 2011, Tropical Storm Irene put the state of Vermont into disaster mode causing widespread damage and destruction. Flood levels approached the historic 1927 Vermont flood. Floodwaters and sediment poured through rivers and communities affecting ~225 municipalities; recovery in Vermont alone is estimated between $700 million and $1 billion. Streambanks hit by these floodwaters tend to fail and erode causing debris and sediment to be suspended in the water. Pollutants bond to the sediment make their way down the rivers draining into Lake Champlain. Identifying the exact source of the suspended sediment is complex, but streambank erosion is thought to be a large factor and is the focus of this study. Aerial photography, a USDA streambank channel model (BSTEM) and a handheld terrestrial LiDAR sensor capable of obtaining high-resolution 3D scans of the streambanks with less than a cm accuracy will be combined to predict the sediment loading from streambank erosion. The goal is to give communities tools they can use to identify potential problem areas along VT streams and make more informed planning decisions.
‘Computers Learn Like Your Biological Neurons’ – Deep-Learning Programs Recognize Patterns in Complex Data
Luke Detwiler (Advisor: Donna M. Rizzo)
Minefields cover large tracts of land in numerous geographic locations across the globe, creating a dangerous hazard for both humans and their environment. A technology called ground-penetrating radar (GPR) is capable of generating images beneath the surface of the earth and shows potential in the detection of buried landmines, provided certain challenges can be overcome (e.g., differentiating between objects such as rocks that share similar properties). This research focuses on the development of a computer algorithm, called a deep-learning Artificial Neural Network (ANN) that currently shows exceptional promise in the field of image recognition. ANNs, in general, work by using an interconnected network of ‘neurons’ that mimic the learning process observed in biological nervous systems. A deep-learning ANN will be coded and trained using ground penetrating radar images to recognize buried foreign objects for future use in landmine detection applications.
Tuff Weathering and the Deteriorations of Cavates
Rebecca Domingue (Advisor: Doug Porter)
Northern New Mexico’s Pajarito Plateau is characterized by the volcanic ash, tuff surface that vary drastically in rock properties. The softest parts of the tuff were extensively carved by the Ancestral Pueblo people who resided in the area to form cavates, or rock-carved room. These cavates are gradually being lost due to weathering processes, such as erosion and exfoliation on the cliff surfaces. However, the cliff surfaces are deteriorating at different rates that seem to correspond with the various rock properties and surface conditions of the tuff. Through tests that measure water uptake, gas permeability and the resistance to surface erosion, the relationship between the processes of deterioration and surface hardness will hopefully be understood. The analysis will provide a better understanding of how to stabilize the cliffs while preserving their historical significance to the Pueblo people.
Effects of the Interstate on Vermont’s Hydrology Cycle
James Laughlin (Advisor: Dr. Paul Bierman, Ana Vang)
Although the construction of the highway systems within Vermont provides access to many places we have never gone before, it may be doing more damage than we can see. Introducing more impervious surface to any given area will change the way water flows, which affects the hydrologic cycle as well as the surrounding landscape. It is important to understand localized impacts to the addition of these surfaces in order to minimize erosion and large-scale landscape changes. This research involves using historic aerial photographs in combination with a two-step computational analysis. The first program, ERDAS Imagine, geometrically corrects the image for distortion caused in the photographic process mainly related to the curvature of the earth. The second program, eCognition, is used to quantify the percent of impervious area both before and after the construction of the highway. The two-step analysis helps identify erosion sensitive areas to help prevent further landscape change.
Comparing Stochastic and Reactive Cellular Approaches to Survival in Fluctuating Environments
James Lent (Advisor: Dr. Mary Dunlop)
In nature organisms face fluctuating environmental conditions to which they must respond in order to survive. Understanding how microbes withstand unfavorable environments has applications ranging from the development of antibiotic treatment regiments to predicting the effects of runoff and oil spills on an ecosystem. This study focuses on two survival strategies in particular: reactive sensory adaptation and random population diversity. This research uses a computer model of E. coli to determine which method is advantageous in each environment. We expect to find that stochastic switching is advantageous when the environment switches suddenly, and with high penalties for being “unfit”, while reactive switching will be better in slow changing environments, or environments with lower penalty for being “unfit”. Our findings provide novel approaches to engineering robust organisms as well as insight into how to eliminate them.
How Do Driving Style and Road Grade Affect Vehicle Tailpipe Emissions?
Anna Nadler (Advisor: Dr. Britt Holmén)
Tailpipe emissions are linked to climate change, contain carcinogens and are associated with cardiovascular disease. The Environmental Protection Agency (EPA) regulates these emissions to reduce the damage of these air pollutants on the environment and human health. The EPA’s MOtor Vehicle Emission Simulator (MOVES) model estimates gas and particulate tailpipe emissions for the on-road vehicle fleet to inform decision-making for new transportation projects. This research compares MOVES model emission rates of 4 regulated gases with real-world, on-road tailpipe emissions data collected by the University of Vermont’s Transportation Air Quality Laboratory (TAQLab). Real-world second-by-second data on vehicle speed, road grade, environmental conditions, fuel, and vehicle type were used as MOVES inputs to estimate tailpipe emissions for a 2010 model passenger car. Preliminary results show good agreement between the real-world and MOVES-pollutant emission rates for downhill road grades; but MOVES overestimates some emission rates for flat and uphill sections. This research will identify the vehicle operating modes associated with the larger emission rate discrepancies (i.e., model versus real-world projections) and investigate the model sensitivity to assumptions and default data to help reduce model errors. Ultimately, model validation, similar to that conducted here, will lead to more accurate vehicle emissions estimates for the late-model passenger vehicles that dominate today’s on-road fleet.
The Structure-Function Relationship of Methacrylated Hyaluronan Hydrogels for Tissue Regeneration
Alex Poniz (Adviser: Dr. Rachael Oldinski)
Osteoarthritis is a degenerative disease affecting cartilage and bone tissues, causing pain and limited mobility. It is the main cause of chronic disability in the United States and affects the productivity and quality of life of millions of people worldwide. Tissue engineering scaffolds are materials (i.e. biomaterials) that cells can grow on, and are promising new tools for regenerating damaged tissue. This summer, I will be testing the physical and mechanical properties of different scaffolds and investigating the relationship between their properties and their ability to regenerate bone and cartilage. The goal of this research is to gain insight into the optimal mechanical properties of scaffolds, how cells react to specific mechanical stimuli, and what makes a scaffold successful.
Sediment Tracing in the Mad River Watershed Through Isotopic Soil Analysis
Marisa Rorabaugh (Advisors: Drs. Donna Rizzo, Mandar Dewoolkar, Kristen Underwood, Scott Hamshaw)
Excess phosphorous and nitrogen within Lake Champlain cause algal blooms that block light, causing death of other plants and a deficiency of dissolved oxygen in the water. These nutrients are carried downstream in the rivers from various point and non-point sources. One of these non-point sources is sediment run-off from various watersheds. The sediment in the Mad River will be traced to its origins within the watershed to identify problem land uses and areas in an attempt to decrease the nutrients traveling to the lake. This will be done using isotopic soil analysis. Most notably, the panel of isotopes will include Delta-13 Carbon (δ13C) and Delta-15 Nitrogen (δ15N), which have never been used as sediment tracers before.
2013 Student Information
Hanna Victoria Anderson will be a junior in the School of Engineering, majoring in environmental engineering. Originally from Vineland, New Jersey she attended Vineland High School. hvanders@uvm.edu
Luke Detwiler will be a senior in the School of Engineering, majoring in environmental engineering. Originally from Natick, Massachusetts, he attended Natick High School.
ldetwile@uvm.edu
Rebecca Domingue will be a senior in the School of Engineering, majoring in civil engineering. Originally from Poland, New York she attended Poland Central School. rrdoming@uvm.edu
James Laughlin will be a senior in the School of Engineering, majoring in environmental engineering. Originally from Massachusetts, he attended Westwood High School. jplaughlin2010@aol.com
James Lent will be a senior in the School of Engineering, majoring in mechanical engineering. Originally from Berkeley, California, he attended Berkeley High School. jdlent@uvm.edu
Anna Elizabeth Nadler will be a senior in the School of Engineering,
majoring in civil engineering. Originally from Rutland, Vermont she attended Rutland High School. aenadler@uvm.edu
Alex Poniz will be a senior in the School of Engineering, majoring in mechanical engineering with a minor in chemistry. Originally from Edison, New Jersey, he attended George School. aponiz@uvm.edu
Marisa Rorabaugh will be a senior in the School of Engineering, majoring in Environmental Engineering. Originally from Fairbanks, Alaska, she attended West Valley High School. mrorabau@uvm.edu
For more information and videos on the Barrett Scholarships visit: http://www.uvm.edu/~cems/soe/?Page=undergrad/ee/barrett/default.php&SM=undergrad/_undergradmenu.html