Project Info
Hydrologic Feature Detection with Python
Matthew Siegfried | siegfried@mines.edu
Our ultimate goal is to detect and categorize polar hydrologic features, and enhance our understanding of the journey, fate, and impact that melting water has on ice dynamics and sea level rise. Our project leverages machine learning techniques that have broad applicability in order to examine rugged, dynamic Arctic environments at the center of modern day climate change.
More Information
Christoffersen, Bougamont, Hubbard, Doyle, Grigsby, and Pettersson, 2018. Cascading lake drainage on the Greenland Ice Sheet triggered by tensile shock and fracture, Nature Communications, 9, https://www.nature.com/articles/s41467-018-03420-8/
Colgan, Rajaram, Abdalati, McCutchan, Mottram, Moussavi, and Grigsby, 2016. Glacier crevasses: Observations, models, and mass balance implications, Reviews of Geophysics, 54(1), https://agupubs.onlinelibrary.wiley.com/doi/full/10.1002/2015RG000504
Stanley, 2016. Recent studies crack open new views of glacial crevasses, Eos, 97, https://eos.org/research-spotlights/recent-studies-crack-open-new-views-of-glacial-crevasses
Witkin, J. 2011. Tracking the Cracks in Greenland’s Ice Sheet. New York Times: Green – Energy, the Environment, and the Bottom Line. https://green.blogs.nytimes.com/2011/09/19/tracking-the-cracks-in-greenlands-ice-sheet/
Grand Engineering Challenge: Not applicable
Student Preparation
Qualifications
Basic understanding of python. Ideally, the student will have a desire to improve the python skills they have and an interest in machine learning. Also, a desire to learn more about scientific programming and the Arctic climate system.
Time Commitment
10-15 hours/week
Skills/Techniques Gained
Software development skills such as working in a collaborative git repository; refinement in python skills; understanding of machine learning practices including model/algorithm selection and optimization, as well as model training, validation, and assessment. This project will also have the opportunity to gain experience with GPU computing, depending on student interest and desire.
Mentoring Plan
The student on this project will be integrated in the mentoring framework of the Mines Glaciology Laboratory. They will be directly supervised by research associate Shane Grigsby, as well as meet weekly with assistant professor Matt Siegfried, who leads the group. Depending on student interest, the student will have the opportunity to work with USGS Hazards group as well, collaborating with Dr. Francis Rengers.
The student will attend fortnightly group meetings with all the undergraduate, graduate, researchers, and faculty members in the group to discuss research progress. The student will present once per semester at this meeting. The student will also have the opportunity to join a reading group for cryosphere research, where peer reviewed literature in the field will be discussed. Based on student desire, the project can culminate in a presentation at a scientific conference, and depending on research progress, S. Grigsby and M. Siegfried can mentor the student in authoring a scientific manuscript for submissions to a peer-reviewed journal.