Glacier Mass Balance Modeling Using a Long Short‐Term Memory Network
Abstract
Glacier mass balance (MB) is a key indicator of climate change and a central driver of glacier evolution, yet most glaciers worldwide lack long‐term in situ measurements. For estimating glacier MB, data‐driven models provide a complementary alternative to traditional numerical approaches by learning empirical relationships between climate forcing, topography, and MB from observations. Here, we develop a recurrent neural network (RNN) based on a Long Short‐Term Memory (LSTM) architecture within the Mass Balance Machine (MBM) framework to predict winter and annual point surface MB across the Swiss Alps. MBM is trained on 30,000 observations from 30 glaciers and tested on eight glaciers excluded from training to assess spatial generalization. MBM predicts winter and annual MB with high accuracy on unseen glaciers (root mean squared error of 0.35 and 0.78 m w.e.). Its recurrent structure enables learning temporal dependencies, improving the representation of seasons with strong accumulation or ablation. Beyond point predictions, MBM generates spatially distributed MB maps that capture MB gradients, and produce glacier‐wide mass changes consistent with geodetic estimates. Monthly outputs further show that MBM reproduces the seasonal transition from winter accumulation to summer ablation with realistic timing and magnitude. These results show that a RNN can recover key characteristics of glacier MB dynamics and that the learned relationships transfer effectively across the climatic and topographic settings of the Swiss Alps. The demonstrated generalization skill highlights the potential of MBM for application in regions with limited direct measurements, though transferability to glaciers with fundamentally different climatic and topographic settings remains to be established.
Related
- Investigation of glacier mass balance based on multi-source observation in the Karakoram over the past 25 years
- Remotely sensed annual glacier mass balance in the China-Pakistan Economic Corridor during 2000–2020: Widespread mass loss and regional contrasts
- Radiation regimes shape microbial light-harvesting strategies and network organization on glacier surfaces
- GlaUnTI: a hybrid physics–machine learning model enables transferable glacier surface mass balance estimation
- Resolving short-term meteorological variability amplifies projected glacier loss and sea-level rise
BibTeX
@article{meer2026glacier,
title = {{Glacier Mass Balance Modeling Using a Long Short‐Term Memory Network}},
author = {Marijn van der Meer and Harry Zekollari and Alban Gossard and Kamilla Hauknes Sjursen and Jordi Bolibar and Matthias Huss and Daniel Farinotti},
journal = {Journal of Geophysical Research. Machine Learning and Computation},
year = {2026},
month = jul,
doi = {10.1029/2026jh001250},
url = {https://hal.science/hal-05856854v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Glacier Mass Balance Modeling Using a Long Short‐Term Memory Network
AU - Marijn van der Meer
AU - Harry Zekollari
AU - Alban Gossard
AU - Kamilla Hauknes Sjursen
AU - Jordi Bolibar
AU - Matthias Huss
AU - Daniel Farinotti
PY - 2026
DA - 2026/07/09
JO - Journal of Geophysical Research. Machine Learning and Computation
DO - 10.1029/2026jh001250
UR - https://hal.science/hal-05856854v1
AB - Glacier mass balance (MB) is a key indicator of climate change and a central driver of glacier evolution, yet most glaciers worldwide lack long‐term in situ measurements. For estimating glacier MB, data‐driven models provide a complementary alternative to traditional numerical approaches by learning empirical relationships between climate forcing, topography, and MB from observations. Here, we develop a recurrent neural network (RNN) based on a Long Short‐Term Memory (LSTM) architecture within the Mass Balance Machine (MBM) framework to predict winter and annual point surface MB across the Swiss Alps. MBM is trained on 30,000 observations from 30 glaciers and tested on eight glaciers excluded from training to assess spatial generalization. MBM predicts winter and annual MB with high accuracy on unseen glaciers (root mean squared error of 0.35 and 0.78 m w.e.). Its recurrent structure enables learning temporal dependencies, improving the representation of seasons with strong accumulation or ablation. Beyond point predictions, MBM generates spatially distributed MB maps that capture MB gradients, and produce glacier‐wide mass changes consistent with geodetic estimates. Monthly outputs further show that MBM reproduces the seasonal transition from winter accumulation to summer ablation with realistic timing and magnitude. These results show that a RNN can recover key characteristics of glacier MB dynamics and that the learned relationships transfer effectively across the climatic and topographic settings of the Swiss Alps. The demonstrated generalization skill highlights the potential of MBM for application in regions with limited direct measurements, though transferability to glaciers with fundamentally different climatic and topographic settings remains to be established.
SN - 2993-5210
ER -