Predicting Glacier Terminus Retreat Using Machine Learning
Creators
- 1. University of Texas Institute for Geophysics (UTIG), Austin, TX, USA
- 2. University of Nevada, Reno, Nevada Seismological Laboratory, Reno, NV, USA
- 3. Cryospheric Sciences Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD, USA
- 4. University of Kansas, Department of Geology, Lawrence, KS, USA
- 5. Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USA
Description
While a majority of mass loss from the Greenland Ice Shelf is attributed to glacial terminus retreat via calving, the superimposed force factors of the ice-ocean interface create a challenge for physically modeling terminus change. Here we use time series of environmental and glacial data, input as features into a machine learning regression model, to forecast terminus retreat for marine-terminating glaciers in Greenland. We then identify the critical features that most impact a glacier’s likelihood of retreat using SHAP analysis. We further analyze the heterogeneous outcomes for individual glaciers to classify them by their terminus change profile. By better understanding the parameters impacting glacial retreat, we inform physical models to reduce uncertainty in mass change projections.
Files
Shionalyn_AGU2022_v3_print.pdf
Files
(2.2 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:f13bc27a66bcce4e63d4784dcb562517
|
2.2 MB | Preview Download |
Additional details
References
- Shekhar et al., 2014, "ALPS: A Unified Framework for Modeling Time Series of Land Ice Changes," Journal of Latex Class Files, v. 13, no. 9.
- Chen et al., 2015, "Xgboost: extreme gradient boosting," R package v. 0.4-2, 1.4.
- Lundberg & Lee, 2017, "A Unified Approach to Interpreting Model Predictions," 31st Conference on Neural Information Processing Systems, Long Central Western Greenland terminus areas Beach, CA, USA.