The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada

Fine-scale maps of ground ice and related surface features are critical for permafrost-related modelling and management. However, such maps are lacking across almost the entire Arctic. Machine learning provides the potential to automate regional fine-scale ground ice mapping using remote sensing and...

Full description

Saved in:
Bibliographic Details
Main Authors: Qianyu Chang, Simon Zwieback, Aaron A. Berg
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/7/1245
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849730375222296576
author Qianyu Chang
Simon Zwieback
Aaron A. Berg
author_facet Qianyu Chang
Simon Zwieback
Aaron A. Berg
author_sort Qianyu Chang
collection DOAJ
description Fine-scale maps of ground ice and related surface features are critical for permafrost-related modelling and management. However, such maps are lacking across almost the entire Arctic. Machine learning provides the potential to automate regional fine-scale ground ice mapping using remote sensing and topographic data. Here, we evaluate the predictive skill of XGBoost models for identifying (1) ice wedge and (2) top-5m visible ground ice in the Tuktoyaktuk Coastlands. We find high predictive skill for ice wedge occurrence (ROC AUC = 0.95, macro F1 = 0.80), with the most important predictors being slope, distance to the coast, and probability of depression. The model accurately predicted regional and local trends in ice wedge occurrence, with an increase in ice wedge polygon (IWP) probability towards the coast and in poorly drained depressions. The model also captured IWP in well-drained uplands of the study area, including locations with poorly visible troughs not contained in the training data. Spatial transferability analyses highlight the regional variability of ice wedge probability, reflecting contrasting climatic and surface conditions. Conversely, the low predictive skill for visible ground ice (ROC AUC = 0.67, macro F1 = 0.53) is attributed to limitations in training data and weak associations with the remotely sensed predictors. The varying predictive accuracy highlights the importance of high-quality reference data and site-specific conditions for improving ground ice studies with data-driven modelling from remote sensing observations.
format Article
id doaj-art-32c8dae3ba8a41ccbb68a39a71bff6c3
institution DOAJ
issn 2072-4292
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-32c8dae3ba8a41ccbb68a39a71bff6c32025-08-20T03:08:54ZengMDPI AGRemote Sensing2072-42922025-04-01177124510.3390/rs17071245The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic CanadaQianyu Chang0Simon Zwieback1Aaron A. Berg2Department of Geography, Environment & Geomatics, University of Guelph, Guelph, ON N1G 2W1, CanadaGeophysical Institute, University of Alaska Fairbanks, Fairbanks, AK 99775, USADepartment of Geography, Environment & Geomatics, University of Guelph, Guelph, ON N1G 2W1, CanadaFine-scale maps of ground ice and related surface features are critical for permafrost-related modelling and management. However, such maps are lacking across almost the entire Arctic. Machine learning provides the potential to automate regional fine-scale ground ice mapping using remote sensing and topographic data. Here, we evaluate the predictive skill of XGBoost models for identifying (1) ice wedge and (2) top-5m visible ground ice in the Tuktoyaktuk Coastlands. We find high predictive skill for ice wedge occurrence (ROC AUC = 0.95, macro F1 = 0.80), with the most important predictors being slope, distance to the coast, and probability of depression. The model accurately predicted regional and local trends in ice wedge occurrence, with an increase in ice wedge polygon (IWP) probability towards the coast and in poorly drained depressions. The model also captured IWP in well-drained uplands of the study area, including locations with poorly visible troughs not contained in the training data. Spatial transferability analyses highlight the regional variability of ice wedge probability, reflecting contrasting climatic and surface conditions. Conversely, the low predictive skill for visible ground ice (ROC AUC = 0.67, macro F1 = 0.53) is attributed to limitations in training data and weak associations with the remotely sensed predictors. The varying predictive accuracy highlights the importance of high-quality reference data and site-specific conditions for improving ground ice studies with data-driven modelling from remote sensing observations.https://www.mdpi.com/2072-4292/17/7/1245ice wedgeremote sensingmachine learningground ice
spellingShingle Qianyu Chang
Simon Zwieback
Aaron A. Berg
The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada
Remote Sensing
ice wedge
remote sensing
machine learning
ground ice
title The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada
title_full The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada
title_fullStr The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada
title_full_unstemmed The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada
title_short The Predictive Skill of a Remote Sensing-Based Machine Learning Model for Ice Wedge and Visible Ground Ice Identification in Western Arctic Canada
title_sort predictive skill of a remote sensing based machine learning model for ice wedge and visible ground ice identification in western arctic canada
topic ice wedge
remote sensing
machine learning
ground ice
url https://www.mdpi.com/2072-4292/17/7/1245
work_keys_str_mv AT qianyuchang thepredictiveskillofaremotesensingbasedmachinelearningmodelforicewedgeandvisiblegroundiceidentificationinwesternarcticcanada
AT simonzwieback thepredictiveskillofaremotesensingbasedmachinelearningmodelforicewedgeandvisiblegroundiceidentificationinwesternarcticcanada
AT aaronaberg thepredictiveskillofaremotesensingbasedmachinelearningmodelforicewedgeandvisiblegroundiceidentificationinwesternarcticcanada
AT qianyuchang predictiveskillofaremotesensingbasedmachinelearningmodelforicewedgeandvisiblegroundiceidentificationinwesternarcticcanada
AT simonzwieback predictiveskillofaremotesensingbasedmachinelearningmodelforicewedgeandvisiblegroundiceidentificationinwesternarcticcanada
AT aaronaberg predictiveskillofaremotesensingbasedmachinelearningmodelforicewedgeandvisiblegroundiceidentificationinwesternarcticcanada