Machine learning predicts global glacier erosion rates with new precision


UVic research predicts worldwide glacier erosion
A glacier in the Canadian Arctic. Credit: John Gosse, Dalhousie University

Glaciers carved the deep valleys of Banff, eroded Ontario to deposit the fertile soils of the Prairies, and continue to change Earth’s surface. But how fast do glaciers sculpt the landscape?

In research published in Nature Geoscience, University of Victoria (UVic) geographer Sophie Norris and her international team provide the most comprehensive view of how fast glaciers erode, and how they change the landscape. Most importantly, their research also provides an estimate of the rate of future erosion for more than 180,000 glaciers worldwide.

Using machine learning-based global analysis, Norris and her research team have worked to predict glacial erosion for 85% of modern glaciers. Their regression equations estimate that 99% of glaciers erode between 0.02 and 2.68 millimeters per year—roughly the width of a credit card.

Machine-learning analysis predicts worldwide glacier erosion
Predicted contemporary glacial erosion rates displayed on a near-global scale. a, Global variability in predicted glacial erosion rates for RGI regions. Owing to incomplete surface velocity and ice thickness data necessary to predict glacial erosion rates, all glaciers in Antarctica (part of RGI region 19) are excluded. bd, Regional variability in predicted glacial erosion rates in the southern Andes (RGI region 17; b), Alaska, western Canada and northwestern USA (RGI zones 1 and 2; c), and South and Central Asia (RGI zones 13, 14 and 15; d). Grid cells are assigned based on the most-frequent erosion rate value, as classified in the key, within each cell (100-km pixels for a; 50-km pixels for bd). Basemaps from Natural Earth (https://www.naturalearthdata.com). Credit: Nature Geoscience (2025). DOI: 10.1038/s41561-025-01747-8

“The conditions that lead to erosion at the base of glaciers are more complicated than we previously understood,” says Norris. “Our analysis found that many variables strongly influence erosion rates: temperature, amount of water under the glacier, what kind of rocks are in the area, and how much heat comes from inside Earth.”

“Given the extreme difficulty in measuring glacial erosion in active glacial settings, this study provides us with estimates of this process for remote locations worldwide,” says John Gosse, Dalhousie University.

Understanding the complex factors that cause erosion underneath glaciers is vital information for landscape management, long-term nuclear waste storage and monitoring the movement of sediment and nutrients around the world.

Norris started this work while a post-doctoral fellow at Dalhousie and concluded it at UVic. The team of collaborators included the University of Grenoble Alpes (France), Dartmouth College (US), Pennsylvania State University (US) and the University of California Irvine (US). The work was carried out in partnership with and financially supported by the Canadian Nuclear Waste Management Organization.

More information:
Sophie L. Norris et al, Drivers of global glacial erosion rates, Nature Geoscience (2025). DOI: 10.1038/s41561-025-01747-8

Citation:
Machine learning predicts global glacier erosion rates with new precision (2025, August 7)
retrieved 8 August 2025
from https://phys.org/news/2025-08-machine-global-glacier-erosion-precision.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.





Source link

Leave a Reply

Translate »
Share via
Copy link