Project Info

Antarctic glacier-ocean interactions using satellite imagery

Matthew Siegfried

Project Goals and Description:

Antarctica may contribute substantially to global sea level rise in the next century, putting coastal communities at greater risk for flooding. The student will be helping to build a training dataset and refine/test cloud-deployable machine learning pipelines that aid in satellite image processing and the detection of ice and ocean features in Antarctica. This work is part of a larger project working to better understand glacier-ocean interactions, which currently drive at least half of the ice loss in Antarctica and are important to understand to improve future projections of sea level change. Tasks may include:
  1. Building a training dataset for Antarctic machine-learning algorithms;
  2. Refining, testing, and deploying a cloud-detection algorithm for Landsat; and
  3. Refining, training, and testing, a machine-learning architecture for ice-front detection.

More Information:

Grand Challenge: Engineer the tools of scientific discovery.
Joughin, I., Alley, R. B. & Holland, D. M. Ice-Sheet Response to Oceanic Forcing. Science 338, 1172–1176 (2012). Alley, K. E., Scambos, T. A., Siegfried, M. R. & Fricker, H. A. Impacts of warm water on Antarctic ice shelf stability through basal channel formation. Nature Geoscience 9, 290–293 (2016). Baumhoer, C. A., Dietz, A. J., Kneisel, C. & Kuenzer, C. Automated Extraction of Antarctic Glacier and Ice Shelf Fronts from Sentinel-1 Imagery Using Deep Learning. Remote Sensing 11, 2529 (2019). Snow, T., Straneo, F., Holte, J., Grigsby, S., Abdalati, W., & Scambos, T. (accepted) More than skin deep: sea surface temperature as a means of inferring Atlantic Water variability on the southeast Greenland continental shelf near Helheim Glacier. Journal of Geophysical Research: Oceans. doi:

Primary Contacts:

Dr. Tasha Snow, | Dr. Matthew Siegfried,

Student Preparation


  • Some computer programming experience, ideally in Python or a willingness to learn Python (helpful but not required)
  • Background knowledge in statistics or machine learning (helpful but not required)




Python proficiency, image processing skills, experience building/training/testing a machine learning algorithm


Students will have regular meetings with Siegfried and Snow to receive feedback and assistance with their project. In those meetings, we will lay out clear goals for the student’s project, track progress, and discuss interesting findings. As is available, the student will also have the ability to write and submit an abstract and give a poster or oral presentation for a relevant conference. Student will have access to regular group meetings with the Mines Glaciology Laboratory to interact with other Mines students more advanced in their degrees (who can also serve as mentors/peers), join paper discussions to gain experience with reading literature, and interact with other guest lecturers in the field.


Share This