Brandon Tober
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Brandon Tober has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
- David Robert Rounce
Department of Civil and Environmental Engineering, Carnegie Mellon University, Pittsburgh, PA, USA
13 shared publications John W. Holt
10 shared publicationsMark Fahnestock
University of Alaska Fairbanks
10 shared publications- Fabien Maussion
Department of Atmospheric and Cryospheric Sciences, University of Innsbruck, Innsbruck, Austria
8 shared publications Christopher F. Larsen
Geophysical Institute, University of Alaska Fairbanks, Fairbanks, AK, USA
7 shared publicationsMartin Truffer
Lunar and Planetary Laboratory University of Arizona Tucson AZ USA
7 shared publications
Latest Publications
- Open Access
- 17 October 2026
- Published by American Geophysical Union (AGU) in Earth's Future
- Vol. 14 (10)
Alaskan Glaciers in a Warming Climate
Glaciers across Alaska are undergoing rapid mass loss, with implications for sea‐level rise, natural hazards, water resources, and more. Here, we utilize new snowline and melt extent observations of Alaskan glaciers at high spatiotemporal resolutions to improve projections to 2100. Specifically, we incorporate these observations into the calibration scheme for the large‐scale Python Glacier Evolution Model (PyGEM), and subsequently simulate future changes under various climate scenarios. We estimate that glaciers in Alaska will lose 49 ± 17% to 73 ± 26% of their 2015 mass by the end of the century, depending on the future climate. Nearly all small glaciers (2 ) will disappear by 2100 regardless of the climate scenario, along with the smallest glacierized subregions of Alaska, namely, the Brooks and Talkeetna ranges. Across larger subregions, negative mass balance is projected to peak in the middle of the 21st century for the modest future warming scenario (i.e., SSP1‐2.6), while negative mass balance continues to increase to 2100 for the most aggressive future warming scenario (SSP5‐8.5). Much of these changes are attributed to increasing temperatures that cause greater melt in addition to a shift in early‐winter precipitation from snow to rain which reduces accumulation and enhances melt by exposing the underlying ice earlier in the next ablation season. Of the remaining glaciers in 2100, 15%–60%, depending on the future climate, have no accumulation area and are thus committed to disappearance. Overall, the widespread loss of glaciers in Alaska by 2100 has implications on ecology, water resources, natural hazards, and tourism across the region.
- 17 October 2026
- Published by American Geophysical Union (AGU) in Earth's Future
- Vol. 14 (10)
Open AccessAlaskan Glaciers in a Warming Climate
Glaciers across Alaska are undergoing rapid mass loss, with implications for sea‐level rise, natural hazards, water resources, and more. Here, we utilize new snowline and melt extent observations of Alaskan glaciers at high spatiotemporal resolutions to improve projections to 2100. Specifically, we incorporate these observations into the calibration scheme for the large‐scale Python Glacier Evolution Model (PyGEM), and subsequently simulate future changes under various climate scenarios. We estimate that glaciers in Alaska will lose 49 ± 17% to 73 ± 26% of their 2015 mass by the end of the century, depending on the future climate. Nearly all small glaciers (2 ) will disappear by 2100 regardless of the climate scenario, along with the smallest glacierized subregions of Alaska, namely, the Brooks and Talkeetna ranges. Across larger subregions, negative mass balance is projected to peak in the middle of the 21st century for the modest future warming scenario (i.e., SSP1‐2.6), while negative mass balance continues to increase to 2100 for the most aggressive future warming scenario (SSP5‐8.5). Much of these changes are attributed to increasing temperatures that cause greater melt in addition to a shift in early‐winter precipitation from snow to rain which reduces accumulation and enhances melt by exposing the underlying ice earlier in the next ablation season. Of the remaining glaciers in 2100, 15%–60%, depending on the future climate, have no accumulation area and are thus committed to disappearance. Overall, the widespread loss of glaciers in Alaska by 2100 has implications on ecology, water resources, natural hazards, and tourism across the region.
- Open Access
- 30 July 2026
- Published by Cambridge University Press (CUP) in Journal of Glaciology
- Vol. 72, 1-35
Multi-decadal observations of glacier thinning from airborne laser altimetry reveal accelerating mass loss in Alaskan National Parks
We use all 27 years of airborne laser altimetry from the University of Alaska Fairbanks to quantify changes in surface elevation across 205 large glaciers that together represent 56% of the glacier area in Alaska and adjacent Canada, including 71% of the ice cover in Alaska’s National Parks. We reprocessed lidar point clouds into a homogenized publicly available 2.5 m digital elevation model (DEM) archive, implemented a new double-differencing approach to incorporate partially overlapping surveys and derived elevation change in 10 m elevation bins to assess glacier-wide mass balance for all surveyed epochs on each glacier. The resulting mean balance rate for all measured glaciers shows statistically significant acceleration of mass loss, ranging from minus 0.83 plus or minus 0.13 mw period e period a minus 1 $-0.83 \pm{0.13} \,\, \text{m}\,\, \text{w.e.}\ \text{a}^{-1}$ before 2008 to minus 1.16 plus or minus 0.08 mw period e period a minus 1 $-1.16 \pm{0.08} \,\, \text{m}\,\, \text{w.e.}\ \text{a}^{-1}$ thereafter. Extrapolating measured thinning rates from the latter period (2008–22) to unsurveyed glaciers suggests that Alaskan National Parks are losing 33.74 Gt a minus 1 $\text{Gt a}^{-1}$ of ice, approximately one third of all estimated losses in the region. We use case studies of several prominent glaciers to show how this sustained acceleration of mass loss, which is consistent with recent global assessments, is having direct impacts on visitors, land managers, ecosystems and geohazards in the National Parks.- 30 July 2026
- Published by Cambridge University Press (CUP) in Journal of Glaciology
- Vol. 72, 1-35
Open AccessMulti-decadal observations of glacier thinning from airborne laser altimetry reveal accelerating mass loss in Alaskan National Parks
We use all 27 years of airborne laser altimetry from the University of Alaska Fairbanks to quantify changes in surface elevation across 205 large glaciers that together represent 56% of the glacier area in Alaska and adjacent Canada, including 71% of the ice cover in Alaska’s National Parks. We reprocessed lidar point clouds into a homogenized publicly available 2.5 m digital elevation model (DEM) archive, implemented a new double-differencing approach to incorporate partially overlapping surveys and derived elevation change in 10 m elevation bins to assess glacier-wide mass balance for all surveyed epochs on each glacier. The resulting mean balance rate for all measured glaciers shows statistically significant acceleration of mass loss, ranging from minus 0.83 plus or minus 0.13 mw period e period a minus 1 $-0.83 \pm{0.13} \,\, \text{m}\,\, \text{w.e.}\ \text{a}^{-1}$ before 2008 to minus 1.16 plus or minus 0.08 mw period e period a minus 1 $-1.16 \pm{0.08} \,\, \text{m}\,\, \text{w.e.}\ \text{a}^{-1}$ thereafter. Extrapolating measured thinning rates from the latter period (2008–22) to unsurveyed glaciers suggests that Alaskan National Parks are losing 33.74 Gt a minus 1 $\text{Gt a}^{-1}$ of ice, approximately one third of all estimated losses in the region. We use case studies of several prominent glaciers to show how this sustained acceleration of mass loss, which is consistent with recent global assessments, is having direct impacts on visitors, land managers, ecosystems and geohazards in the National Parks. - Open Access
- 24 July 2026
- Published by Cambridge University Press (CUP) in Journal of Glaciology
- Vol. 72, 1-22
Projections of 21st-century glacier change in Alaska’s national parks calibrated against spatially distributed elevation change observations
Despite advances in global glacier modeling, notable uncertainties remain in part due to model overparameterization. We develop a novel Bayesian inference framework to calibrate the Python Glacier Evolution Model (PyGEM) against spatially distributed surface elevation change data. The utility of this framework is demonstrated using a 1994–2021 airborne laser altimetry record for 185 glaciers in Alaska, which altogether represent over half of the region’s glacier area. Calibration against these data reduces parameter uncertainty by 5–18% and lowers the mean absolute error in hindcast surface elevation change by 0.18 m a −1 (15% compared to prior models). Model projections for these 185 glaciers estimate mass loss ranging from 43 plus or minus $ \pm $ 17% to 67 plus or minus $ \pm $ 27% depending on the emissions scenario. The projections highlight considerable subregional variability ranging from a median glacier mass loss of 57–99% by the end of the century for Lake Clark, Denali, Kenai Fjords, and Glacier Bay national parks, compared to 28–51% lost for Wrangell–St. Elias National Park and Preserve. As additional spatially distributed elevation change datasets become available in Alaska and elsewhere, the framework presented here is well suited to further refine model calibration and improve projections, while remaining flexible to incorporate additional glacier change observations as they become available.- 24 July 2026
- Published by Cambridge University Press (CUP) in Journal of Glaciology
- Vol. 72, 1-22
Open AccessProjections of 21st-century glacier change in Alaska’s national parks calibrated against spatially distributed elevation change observations
Despite advances in global glacier modeling, notable uncertainties remain in part due to model overparameterization. We develop a novel Bayesian inference framework to calibrate the Python Glacier Evolution Model (PyGEM) against spatially distributed surface elevation change data. The utility of this framework is demonstrated using a 1994–2021 airborne laser altimetry record for 185 glaciers in Alaska, which altogether represent over half of the region’s glacier area. Calibration against these data reduces parameter uncertainty by 5–18% and lowers the mean absolute error in hindcast surface elevation change by 0.18 m a −1 (15% compared to prior models). Model projections for these 185 glaciers estimate mass loss ranging from 43 plus or minus $ \pm $ 17% to 67 plus or minus $ \pm $ 27% depending on the emissions scenario. The projections highlight considerable subregional variability ranging from a median glacier mass loss of 57–99% by the end of the century for Lake Clark, Denali, Kenai Fjords, and Glacier Bay national parks, compared to 28–51% lost for Wrangell–St. Elias National Park and Preserve. As additional spatially distributed elevation change datasets become available in Alaska and elsewhere, the framework presented here is well suited to further refine model calibration and improve projections, while remaining flexible to incorporate additional glacier change observations as they become available. - Open Access
- 20 June 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Earth Surface
- Vol. 131 (6)
Alaska‐Yukon Glacier Depths From a Decade of Airborne Radar Sounding
NASA's Operation IceBridge employed airborne radar sounders in Alaska and adjacent northwestern Canada between 2012 and 2021 to measure the thickness of the region's glaciers. Here we present the first comprehensive analysis of these data, analyzing ∼31,700 linear‐km of radar profile data to provide over 5,500 linear‐km of ice thickness and bed elevation measurements from 88 glaciers—constituting the greatest ice thickness inventory for this region to date. Aside from glaciers of the Saint Elias Mountains, radar bed returns are limited to expansive accumulation areas and glacier termini, distant from sources of off‐nadir surface topography. Gridded measurements across Bering Glacier reveal a subglacial trough extending over 50 km from the glacier's terminus up to the Bagley Ice Valley, likely a subglacial expression of the Bering Fault. We find that many of the glacier termini successfully sounded by Operation IceBridge have overdeepened beds, which may offer insight into the potential extent of proglacial lakes and associated natural hazards given continued thinning and retreat. While the long‐wavelength sounders employed by Operation IceBridge have proven capable of sounding through nearly 1,500 m of temperate ice, radar surface returns from the flanks of the region's mountain glaciers remain the greatest challenge to identifying glacier bed returns and retrieving ice thickness measurements. Simulating and accounting for these returns in survey planning may significantly improve the mapping success of future airborne radar campaigns.
- 20 June 2026
- Published by American Geophysical Union (AGU) in Journal of Geophysical Research: Earth Surface
- Vol. 131 (6)
Open AccessAlaska‐Yukon Glacier Depths From a Decade of Airborne Radar Sounding
NASA's Operation IceBridge employed airborne radar sounders in Alaska and adjacent northwestern Canada between 2012 and 2021 to measure the thickness of the region's glaciers. Here we present the first comprehensive analysis of these data, analyzing ∼31,700 linear‐km of radar profile data to provide over 5,500 linear‐km of ice thickness and bed elevation measurements from 88 glaciers—constituting the greatest ice thickness inventory for this region to date. Aside from glaciers of the Saint Elias Mountains, radar bed returns are limited to expansive accumulation areas and glacier termini, distant from sources of off‐nadir surface topography. Gridded measurements across Bering Glacier reveal a subglacial trough extending over 50 km from the glacier's terminus up to the Bagley Ice Valley, likely a subglacial expression of the Bering Fault. We find that many of the glacier termini successfully sounded by Operation IceBridge have overdeepened beds, which may offer insight into the potential extent of proglacial lakes and associated natural hazards given continued thinning and retreat. While the long‐wavelength sounders employed by Operation IceBridge have proven capable of sounding through nearly 1,500 m of temperate ice, radar surface returns from the flanks of the region's mountain glaciers remain the greatest challenge to identifying glacier bed returns and retrieving ice thickness measurements. Simulating and accounting for these returns in survey planning may significantly improve the mapping success of future airborne radar campaigns.
- Open Access
- 14 March 2026
- Published by Copernicus GmbH
Probabilistic calibration of glacier projections using data assimilation
Global glacier mass loss is accelerating, yet considerable spread in projected glacier changes remains due to model structure, forcing, and parameter uncertainty. Most global glacier projections rely on relatively simple models where calibration approaches are essential to constrain uncertainty by using scarce observations. Probabilistic calibration strategies that help with quantifying uncertainities have recently incorporated into a limited number of global glacier models. As a next step, we suggest employing Bayesian data assimilation methods with untapped potential to integrate multi source observations (in-situ, satellite, reanalysis) when calibrating global glacier model projections.Here, we present a probabilistic calibration framework for the Python Glacier Evolution Model (PyGEM) based on ensemble-based data assimilation. Two data assimilation techniques (PBS and AdaPBS) are used to calibrate mass balance parameters to constrain future projections of glacio-hydrological variables (surface mass balance and runoff) between 2015 and 2100 under four SSP climate scenarios across the Scandinavian region. In this work, we compare our calibration results with commonly used deterministic and probabilistic glacier model calibration algorithms which are already in use in PyGEM. Our probabilistic calibration framework based on data assimilation has the potential to quantify and disentangle uncertainties from climate forcing, model structure, and parameters. Better constrained and uncertainty-aware glacier models increases confidence in projections of future glacier change and their relevant impacts. This work paves the way for producing policy relevant global glacier projections and scenarios with their uncertainty estimates within the ERC-AdG GLACMASS project.
- 14 March 2026
- Published by Copernicus GmbH
Open AccessProbabilistic calibration of glacier projections using data assimilation
Global glacier mass loss is accelerating, yet considerable spread in projected glacier changes remains due to model structure, forcing, and parameter uncertainty. Most global glacier projections rely on relatively simple models where calibration approaches are essential to constrain uncertainty by using scarce observations. Probabilistic calibration strategies that help with quantifying uncertainities have recently incorporated into a limited number of global glacier models. As a next step, we suggest employing Bayesian data assimilation methods with untapped potential to integrate multi source observations (in-situ, satellite, reanalysis) when calibrating global glacier model projections.Here, we present a probabilistic calibration framework for the Python Glacier Evolution Model (PyGEM) based on ensemble-based data assimilation. Two data assimilation techniques (PBS and AdaPBS) are used to calibrate mass balance parameters to constrain future projections of glacio-hydrological variables (surface mass balance and runoff) between 2015 and 2100 under four SSP climate scenarios across the Scandinavian region. In this work, we compare our calibration results with commonly used deterministic and probabilistic glacier model calibration algorithms which are already in use in PyGEM. Our probabilistic calibration framework based on data assimilation has the potential to quantify and disentangle uncertainties from climate forcing, model structure, and parameters. Better constrained and uncertainty-aware glacier models increases confidence in projections of future glacier change and their relevant impacts. This work paves the way for producing policy relevant global glacier projections and scenarios with their uncertainty estimates within the ERC-AdG GLACMASS project.
- Open Access
- 13 March 2026
- Published by Copernicus GmbH
Present and future spatiotemporal patterns of glacier change across Alaska
Mountain glaciers affect water resources, natural hazards, sea-level rise, and more. Snowlines and melt extents are surrogate measurements of glacier mass balance that improve our understanding of glacier changes and help constrain models. Sentinel-1 synthetic aperture radar (SAR) provides reliable data every 12 days since ~2016 regardless of cloud coverage or daylight. We present a framework to derive glacier melt extents and transient snowlines from SAR data and demonstrate its utility for 99% of glaciers with an area of at least 2 km2 in Alaska, representing over 85% of the total glaciated area in Alaska. We subsequently leverage these observations as calibration data for the large-scale Python Glacier Evolution Model (PyGEM) and quantify changes in mass and equilibrium line altitude through the end of the 21st century for various emissions scenarios. We highlight stark spatial and temporal patterns in melt across and within various subregions in the observational dataset, as well as long-term trends revealed via modeling. All of our results are produced via an automated workflow that can easily be applied to other regions across the world.
- 13 March 2026
- Published by Copernicus GmbH
Open AccessPresent and future spatiotemporal patterns of glacier change across Alaska
Mountain glaciers affect water resources, natural hazards, sea-level rise, and more. Snowlines and melt extents are surrogate measurements of glacier mass balance that improve our understanding of glacier changes and help constrain models. Sentinel-1 synthetic aperture radar (SAR) provides reliable data every 12 days since ~2016 regardless of cloud coverage or daylight. We present a framework to derive glacier melt extents and transient snowlines from SAR data and demonstrate its utility for 99% of glaciers with an area of at least 2 km2 in Alaska, representing over 85% of the total glaciated area in Alaska. We subsequently leverage these observations as calibration data for the large-scale Python Glacier Evolution Model (PyGEM) and quantify changes in mass and equilibrium line altitude through the end of the 21st century for various emissions scenarios. We highlight stark spatial and temporal patterns in melt across and within various subregions in the observational dataset, as well as long-term trends revealed via modeling. All of our results are produced via an automated workflow that can easily be applied to other regions across the world.
- Open Access
- 13 March 2026
- Published by Copernicus GmbH
Peak glacier extinction in the mid-twenty-first century
Recognising the global importance and vulnerability of mountain glaciers, the United Nations General Assembly declared 2025 the International Year of Glaciers’ Preservation, highlighting the critical role of glaciers in the hydrological cycle and the growing societal risks associated with their rapid decline. These concerns are well founded: glaciers worldwide are retreating rapidly, yet projections have traditionally focused on changes in mass and area rather than on the fate of individual glaciers.Here, we quantify the future evolution of the global number of glaciers and introduce the concept of peak glacier extinction: the year in which the largest number of individual glaciers is projected to disappear. Using three global glacier models and the Randolph Glacier Inventory v6.0, we project the fate of more than 200,000 glaciers worldwide under four policy-relevant global warming scenarios by 2100 (+1.5 °C, +2.0 °C, +2.7 °C and +4.0 °C relative to pre-industrial levels). A glacier is classified as extinct when its area falls below 0.01 km² or its remaining volume declines to less than 1% of its initial value.Across all scenarios, we identify a pronounced mid-century peak in glacier extinction. Under a +1.5 °C pathway, this peak occurs around 2041, with approximately 2,000 glaciers disappearing per year, whereas under a +4.0 °C scenario it shifts to the mid-2050s and intensifies to nearly 4,000 glacier extinctions annually. Regional differences reflect contrasts in glacier size distributions and climatic conditions. Our results highlight the urgency of ambitious climate action. Whether the world faces the loss of 2,000 or 4,000 glaciers per year by mid-century, and whether roughly 20,000 or 100,000 glaciers remain by the end of the century, will be determined by near-term policy and societal choices made today.
- 13 March 2026
- Published by Copernicus GmbH
Open AccessPeak glacier extinction in the mid-twenty-first century
Recognising the global importance and vulnerability of mountain glaciers, the United Nations General Assembly declared 2025 the International Year of Glaciers’ Preservation, highlighting the critical role of glaciers in the hydrological cycle and the growing societal risks associated with their rapid decline. These concerns are well founded: glaciers worldwide are retreating rapidly, yet projections have traditionally focused on changes in mass and area rather than on the fate of individual glaciers.Here, we quantify the future evolution of the global number of glaciers and introduce the concept of peak glacier extinction: the year in which the largest number of individual glaciers is projected to disappear. Using three global glacier models and the Randolph Glacier Inventory v6.0, we project the fate of more than 200,000 glaciers worldwide under four policy-relevant global warming scenarios by 2100 (+1.5 °C, +2.0 °C, +2.7 °C and +4.0 °C relative to pre-industrial levels). A glacier is classified as extinct when its area falls below 0.01 km² or its remaining volume declines to less than 1% of its initial value.Across all scenarios, we identify a pronounced mid-century peak in glacier extinction. Under a +1.5 °C pathway, this peak occurs around 2041, with approximately 2,000 glaciers disappearing per year, whereas under a +4.0 °C scenario it shifts to the mid-2050s and intensifies to nearly 4,000 glacier extinctions annually. Regional differences reflect contrasts in glacier size distributions and climatic conditions. Our results highlight the urgency of ambitious climate action. Whether the world faces the loss of 2,000 or 4,000 glaciers per year by mid-century, and whether roughly 20,000 or 100,000 glaciers remain by the end of the century, will be determined by near-term policy and societal choices made today.
- Open Access
- 15 December 2025
- Published by Springer Nature in Nature Climate Change
- Vol. 16 (2) , 143-147
Peak glacier extinction in the mid-twenty-first century
Projections of glacier change typically focus on mass and area loss, yet the disappearance of individual glaciers directly threatens culturally, spiritually and touristically significant landscapes. Here, using three global glacier models, we project a sharp rise in the number of glaciers disappearing worldwide, peaking between 2041 and 2055 with up to ~4,000 glaciers vanishing annually. Regional variability reflects differences in average glacier size, local climate, the magnitude of warming and inventory completeness.
- 15 December 2025
- Published by Springer Nature in Nature Climate Change
- Vol. 16 (2) , 143-147
Open AccessPeak glacier extinction in the mid-twenty-first century
Projections of glacier change typically focus on mass and area loss, yet the disappearance of individual glaciers directly threatens culturally, spiritually and touristically significant landscapes. Here, using three global glacier models, we project a sharp rise in the number of glaciers disappearing worldwide, peaking between 2041 and 2055 with up to ~4,000 glaciers vanishing annually. Regional variability reflects differences in average glacier size, local climate, the magnitude of warming and inventory completeness.
- Open Access
- 17 November 2025
- Published by Copernicus GmbH in The Cryosphere
- Vol. 19 (11) , 5801-5826
Machine learning improves seasonal mass balance prediction for unmonitored glaciers
Glacier evolution models based on temperature-index approaches are commonly used to assess hydrological impacts of glacier changes. However, current model calibration frameworks cannot efficiently transfer information from sparse high-resolution observations across glaciers. This limits their ability to resolve seasonal mass changes on unmonitored glaciers in large-scale applications. Machine learning approaches can potentially address this limitation by learning relationships from sparse data that are transferable in space and time, including to unmonitored glaciers. Here, we present the Mass Balance Machine (MBM), a data-driven mass balance model based on the XGBoost architecture, designed to provide accurate and high spatio-temporal resolution regional-scale reconstructions of glacier mass balance. We trained and tested MBM using a dataset of approximately 4000 seasonal and annual point mass balance measurements from 32 glaciers across heterogeneous climate settings in mainland Norway, spanning from 1962 to 2021. To assess the advantage of MBM's generalisation capabilities, we compared its predictions on independent test glaciers at various spatio-temporal scales with those of regional-scale simulations from three glacier evolution models. MBM successfully predicted annual and seasonal point mass balance on the test glaciers (RMSE of 0.59–1.00 m w.e. and bias of −0.01 to 0.04 m w.e.). On seasonal mass balance, MBM outperformed the other models across spatial scales, reducing RMSE by up to 46 % and 25 % on glacier-wide winter and summer mass balance, respectively. Our results demonstrate the capability of machine learning models to generalise across glaciers and climatic settings from relatively sparse mass balance data, highlighting their potential for a wide range of applications.
- 17 November 2025
- Published by Copernicus GmbH in The Cryosphere
- Vol. 19 (11) , 5801-5826
Open AccessMachine learning improves seasonal mass balance prediction for unmonitored glaciers
Glacier evolution models based on temperature-index approaches are commonly used to assess hydrological impacts of glacier changes. However, current model calibration frameworks cannot efficiently transfer information from sparse high-resolution observations across glaciers. This limits their ability to resolve seasonal mass changes on unmonitored glaciers in large-scale applications. Machine learning approaches can potentially address this limitation by learning relationships from sparse data that are transferable in space and time, including to unmonitored glaciers. Here, we present the Mass Balance Machine (MBM), a data-driven mass balance model based on the XGBoost architecture, designed to provide accurate and high spatio-temporal resolution regional-scale reconstructions of glacier mass balance. We trained and tested MBM using a dataset of approximately 4000 seasonal and annual point mass balance measurements from 32 glaciers across heterogeneous climate settings in mainland Norway, spanning from 1962 to 2021. To assess the advantage of MBM's generalisation capabilities, we compared its predictions on independent test glaciers at various spatio-temporal scales with those of regional-scale simulations from three glacier evolution models. MBM successfully predicted annual and seasonal point mass balance on the test glaciers (RMSE of 0.59–1.00 m w.e. and bias of −0.01 to 0.04 m w.e.). On seasonal mass balance, MBM outperformed the other models across spatial scales, reducing RMSE by up to 46 % and 25 % on glacier-wide winter and summer mass balance, respectively. Our results demonstrate the capability of machine learning models to generalise across glaciers and climatic settings from relatively sparse mass balance data, highlighting their potential for a wide range of applications.
- Open Access
- 13 October 2025
- Published by Cambridge University Press (CUP) in Journal of Glaciology
- Vol. 71
A century of flow and surge history of Sít’ Tlein (Malaspina Glacier), Southeast Alaska – ERRATUM
- 13 October 2025
- Published by Cambridge University Press (CUP) in Journal of Glaciology
- Vol. 71
Open AccessA century of flow and surge history of Sít’ Tlein (Malaspina Glacier), Southeast Alaska – ERRATUM