Environmental Internship Program - 2023 Booklet
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
WATER AND THE<br />
ENVIRONMENT<br />
PROJECT TITLE<br />
Understanding Watershed<br />
Processes in Complex<br />
Terrain – Mountain<br />
Hydrology Field Camp<br />
ORGANIZATION(S)<br />
Integrated GroundWater<br />
Modeling Center,<br />
Department of Civil<br />
and <strong>Environmental</strong><br />
Engineering, Princeton<br />
University<br />
LOCATION(S)<br />
Rocky Mountain<br />
Biological Laboratory,<br />
Gothic, Colorado<br />
MENTOR(S)<br />
Reed Maxwell,<br />
William and Edna<br />
Macaleer Professor of<br />
Engineering and Applied<br />
Science, Professor of Civil<br />
and <strong>Environmental</strong><br />
Engineering and the High<br />
Meadows <strong>Environmental</strong><br />
Institute; Harry Stone,<br />
Ph.D. candidate, Civil and<br />
<strong>Environmental</strong><br />
Engineering<br />
Sarah Burbank ’25<br />
COMPUTER SCIENCE<br />
Certificates: African American Studies,<br />
Quantitative and Computational Biology<br />
Mountainous watersheds are important for<br />
recharging the flow of Western American rivers.<br />
Understanding these watersheds is critical<br />
to modeling how climate change will affect<br />
water resources, however, they are difficult to<br />
model due to their complex terrain. I aimed<br />
to improve understanding of the spatial and<br />
temporal factors that govern soil moisture in<br />
mountainous watersheds and to investigate the<br />
viability of using machine learning to predict<br />
soil moisture in highly complex terrain from<br />
in situ measurements. I helped collect various<br />
data over a small drainage in Colorado’s East<br />
River watershed, including soil moisture data,<br />
meteorological data and drone-collected<br />
topographical characterization data. Then,<br />
I used these datasets to run a random forest<br />
regression machine-learning model to predict<br />
soil moisture. These methods can be used to<br />
extrapolate the findings of labor-intensive field<br />
campaigns to larger areas. I learned skills in<br />
data collection, organization and preprocessing.<br />
Seeing how research can translate data from<br />
real environmental trends into a computational<br />
output was exciting. The experience also gave<br />
me the opportunity to interact with successful<br />
people at all levels of academia, which enabled<br />
me to envision a future career path doing the<br />
same.<br />
96