Willem Jan van de Berg

Assistent Professor

Utrecht University
Country flagNetherlands
ApplyKite TwinLearns as you apply

Make your next move.

Prepare your application

Review drafts before you send them.

Explore related searches

Articles (37)

Recent observations reveal that the West Antarctic Ice Sheet is rapidly thinning, particularly at its two largest outlet glaciers, Pine Island Glacier and Thwaites Glacier, while East Antarctica remains relatively stable. Ice sheet model projections over the next few centuries give a mixed picture, some ice sheet models forced by climate models project mass gain by increased surface mass balance, while most models project moderate to severe mass loss by increasing ice discharge. In this study, we explore the effect of present-day ice thickness change rates on forced future simulations of the Antarctic Ice Sheet using the Community Ice Sheet Model (CISM). We start with a series of schematic, uniform ocean temperature perturbations in the Amundsen Sea Embayment (ASE) to probe the sensitivity of the modelled present-day imbalance to ocean warming. We then apply ocean and atmospheric forcing from seven datasets produced by five Earth System Models (ESMs) from the CMIP5 and CMIP6 ensemble to simulate the Antarctic Ice Sheet from 2015 to 2500. The schematic experiments suggest the presence of an ice-dynamical limit; Thwaites Glacier (TG) does not collapse in these experiments (i.e., with accelerated deglaciation leading to considerable grounded ice mass loss) before ∼2100 without more than 2° of ocean warming. Meanwhile, the maximum rate of Global Mean Sea Level rise (GMSLr) from the ASE during the collapse increases linearly with ocean temperature, indicating that while earlier collapse timing shows diminishing returns, the rate of sea-level rise keeps on intensifying with stronger forcing. The relative importance of including the observed present-day mass loss rates decreases for larger (ocean) warming under climate forcing, and decreases over time. For the East Antarctic Ice Sheet on shorter timescales (until 2100), adding the present-day observed mass change rates doubles its global mean sea level rise contribution. On longer timescales (2100–2500), the effect of the present-day observed mass change rates is smaller. Thinning of the West Antarctic Ice Sheet induced by the present-day imbalance is partly compensated by present-day ice sheet thickening of the East Antarctic Ice Sheet over the coming centuries, which persists in our simulations. These deviations are overshadowed by the mass losses induced by the projected ocean warming.

Year:

2026

Three recent downscalings of CESM with MAR, RACMO, and HIRHAM under SSP5-8.5 produce consistent contemporary Antarctic surface mass balance (SMB), but diverge strongly by 2100, especially over ice shelves. HIRHAM simulates a large SMB decline driven by strong runoff increases, MAR a moderate decrease, while RACMO maintains near balance. These differences mainly reflect contrasting melt–albedo feedbacks, present-day melt and refreezing levels, and a persistent 1–2 °C temperature offset between MAR and RACMO. CESM shows a decline similar in magnitude to MAR, with high melt partly compensated by extensive refreezing. Over grounded ice, all models project increased SMB from higher snowfall, though runoff still drives their spread. Despite shared boundary conditions and similar contemporary SMB, model behavior diverges, and CESM’s integrated results resemble MAR's despite its coarse resolution. Performance differences in present-day melt suggest that uncertainty estimates should account for model skill, motivating Bayesian treatment of multi-model ensembles.

Year:

2026

The firn layer regulates how an ice-sheet responds to climate change by modifying how changes in surface temperature, snow accumulation and ablation affect the ice-sheet mass balance. Firn properties are often simulated with a firn densification model. Firn models are forced with surface mass balance components, surface energy balance components and/or meteorological variables. In literature, a variety of climate forcing time steps have been used to force such firn models, ranging from 3-hours to 1-day, 1-month, or even annual. To investigate the impact of these different time steps, we force the firn densification model IMAU-FDM with different climate forcing time steps at the surface for the Antarctic Peninsula and southern Greenland Ice Sheet. We show that the modelled firn layer contains more pore space for larger forcing time steps, and that locations with limited firn pore space due to seasonal melt are most sensitive. The climate forcing time step impacts the creation of pore space by snowfall and the depletion of pore space by snowmelt and firn densification. The key in causing the differences in firn pore space is the presence or absence of a diurnal cycle in the input data. A climate forcing time step greater than a day allows for a non-physical coexistence of snowmelt and sub-zero surface temperatures, leading to immediate shallow refreezing of meltwater. Subsequent melting removes refrozen ice rather than porous firn, reducing the amount of firn air that is lost through melting. Therefore, the decoupled temperature and snowmelt in the upper layers give more firn air with a climate forcing time step larger than a day. We also found that firn density parameterizations can become unsuitable when applied outside the physical conditions or climate forcing time step on which they are based. These parameterizations lead to unrealistic firn densification and accumulation behavior in the model. We argue that (1) firn models require a timestep small enough to capture at least the diurnal cycle, (2) use of parameterizations should be critically assessed and used consistently with the way they were originally developed, and (3) the forcing time step should be considered when interpreting firn model output.

Year:

2026

The Greenland ice sheet (GrIS) is a major contributor to global sea level rise. A significant portion of the GrIS’ contribution can be attributed to increased ice surface melting, which is strongly controlled by ice albedo, a property that regulates the amount of absorbed solar radiation that leads to surface melting. Yet, we lack a comprehensive understanding of the complex and nonlinear relationships ice albedo has with its environment and is, therefore, often simplified or crudely parameterized in climate models. However, an accurate representation of future ice albedo evolution is essential for reducing uncertainties in sea level rise projections. This study presents PIXAL, a physics-informed explainable machine learning architecture that significantly outperforms the Modèle Atmosphérique Régional (MAR), a state-of-the-art regional climate model, in modeling ice albedo on the southwestern GrIS. PIXAL is based on an Extreme Gradient Boosting (XGBoost) model and is trained on a suite of modeled topographic, atmospheric, radiative, and glaciologic variables from MAR to capture the complex and nonlinear relationships with ice albedo observations obtained from the Moderate Resolution Imaging Spectroradiometer (MODIS). Performance metrics show that PIXAL achieves an R2 of 0.563, a structural similarity index measure (SSIM) of 0.620, a mean squared error (MSE) of 0.005, and a mean absolute percentage error (MAPE) of 14.699 %, compared to MAR’s R2 of 0.062, SSIM of 0.112, MSE of 0.032, and MAPE of 46.202 %. Explainable artificial intelligence analysis reveals that topographic features, specifically ice sheet surface height and slope, are the most important drivers of ice albedo variability due to their relationships with ice exposure duration and the effectiveness in accumulating meltwater and light-absorbing constituents (LACs) on flat ice surfaces. Near-surface air temperature and runoff further significantly impact ice albedo variability by affecting the ice metamorphic state and accumulation of meltwater and LACs. These findings highlight that understanding the complex physical processes underlying ice albedo variability is essential for accurate climate modeling and sea level rise predictions. PIXAL represents a crucial advancement in ice albedo modeling and paves the way for improved representation of ice sheets in Earth system models.

Year:

2026

Previous studies do not agree on the magnitude of the influence of basal friction laws in sea-level projections. We use the Community Ice Sheet Model (CISM) to show that the sensitivity of the projected sea level rise to the choice of basal friction law depends on the specific geometric setting and the initial state of the ice sheet model. We find a geometry-driven connection between buttressing and basal sliding in the Amundsen Sea Embayment when performing multi-century future simulations based on the present-day observed imbalance of the Antarctic Ice Sheet, in which Thwaites and Pine Island glaciers eventually collapse. We perform two initializations which differ in their value of a free parameter that governs the effective pressure of ice grounded on bedrock below sea level. Both initializations lead to a modelled Antarctic Ice Sheet that well resembles present-day conditions (ice thickness, ice surface velocities and mass changes rates). Following each initialization, we run the model forward with present-day climate forcing. In one simulation, Thwaites Glacier collapses first, and in the other, Pine Island Glacier collapses first. When Thwaites Glacier collapses first, it creates a grounding line flux large enough to sustain an ice shelf that provides buttressing which largely balances the basal friction differences when using different basal friction parameterizations. A collapsing Pine Island Glacier, however, is sensitive to the choice of basal friction law. Thus, the regional evolution and its sensitivity to friction laws depend on initialization choices that are poorly constrained by observations. In both simulations, present-day ocean thermal forcing, which is a product of the inversion using the present-day imbalance, is sufficient to drive Thwaites and Pine Island collapse, but small differences introduce an uncertainty of about 500 years in the collapse timing.

Year:

2026

Clouds present one of the major challenges for polar climate modeling and significantly contribute to uncertainties in climate and ice sheet mass balance projections, as their radiative effect can strongly impact ice and snow melt. Therefore, a reliable representation of clouds in polar climate models is essential, yet the observations necessary for their evaluation remain sparse. The launch of the Earth Cloud, Aerosol, and Radiation Explorer (EarthCARE) satellite in May 2024 helps bridge this gap by offering cloud observations in unprecedented detail using multiple instruments. Here, we demonstrate the potential of using these novel observations to evaluate cloud representation over the Greenland ice sheet in the regional climate model RACMO (version 2.4p1). To this end, we show along-track comparisons of co-located RACMO cloud profiles with EarthCARE lidar and radar observations. We compare both lidar backscatter and radar reflectivity observations, as well as retrieved cloud properties, with simulated RACMO profiles for two selected case studies. These first results indicate that RACMO simulates low- and mid-altitude ice clouds and snowfall at the correct locations, but fails to capture thinner high-altitude clouds. Additionally, RACMO typically underestimates cloud ice and snow water content, in particular in precipitating systems, where RACMO underestimates snowfall rates. Regarding supercooled liquid and mixed-phase clouds, RACMO does not always reproduce these, especially when they are located at higher altitudes. These first comparisons highlight the potential for using EarthCARE observations to evaluate regional climate models and provide directions for further development of RACMO.

Year:

2025

Antarctica's weather and climate have global impacts, influencing weather patterns, ocean currents and sea levels worldwide. However, Antarctica is vast and complex, and the atmospheric processes that govern its climate are strongly influenced by its steep terrain, particularly around the coastal periphery. Our scientific understanding of this complex environment is hampered by the lack of reliable observations and gridded datasets at sufficiently high spatial and temporal resolution. High-resolution regional climate models, RCMs, can provide a solution to the sparsity of observational data and low resolution of reanalyses, facilitating more in-depth assessments of crucial climate variables like precipitation, winds and temperatures that are strongly influenced by topography. Here we present and evaluate a comprehensive, high-quality, ~ 11 km resolution RCM dataset: the PolarRES ensemble. We show that the ensemble largely out-performs ERA5, particularly for variables such as coastal winds and in characterising high-resolution regional precipitation patterns. There are no consistent seasonal differences in biases, but there are persistent regional biases. The Victoria Land region is the region the RCMs and ERA5 struggle the most with, which suggests that further investigation and model development is needed in this area. Each RCM has strengths and limitations, but overall the ensemble captures the observed weather and climate of Antarctica well. The PolarRES ensemble offers a novel and exciting way of evaluating climate processes and features, and we encourage researchers to use the data, which are freely available, to explore pertinent climate questions of local, regional and global significance.

Year:

2025

Surface meltwater can saturate firn, form melt ponds, and trigger hydrofracturing of Antarctic ice shelves, ultimately accelerating grounded ice flow and contributing to sea level rise. Although the response of surface melt to atmospheric warming (expressed by near-surface air temperature) is known to be non-linear, the mechanisms driving this non-linearity remain poorly understood. In this study we explain the non-linear temperature-melt relationship from an energy balance perspective and assess its spatial variability across Antarctic ice shelves. We use the regional climate model RACMO2.4p1, forced by ERA5 re-analysis and two global earth system models under the SSP3-7.0 high emission scenario, to simulate contemporary and future Antarctic climate and surface mass balance until 2100. We find that the temperature dependence of net shortwave radiation is the primary driver of the non-linearity. On relatively cold ice shelves, warming increases cloud cover and snowfall, raising albedo, reducing net shortwave radiation. In contrast, on warmer ice shelves the snowmelt-albedo feedback dominates the response: warming leads to melt that reduces albedo, enhancing shortwave radiation absorption. The temperature–melt relationship also varies spatially: ice shelves in drier regions experience more melt at the same average summer temperatures than those in wetter regions, highlighting the role of snowfall in suppressing the albedo feedback. When mean summer air temperatures reach or exceed the melting point (0 °C), ice shelves becomes even more sensitive to warming. Surface temperatures can not rise above 0 °C while the atmosphere can, allowing the sensible heat and net longwave radiation to increase. At the same time, snowfall transitions to rain, amplifying the albedo feedback. Our results suggest that currently colder, drier and stable ice shelves could experience rapid increases in melt under future warming, with implications for their long-term stability.

Year:

2025

Recent observations reveal that the West Antarctic Ice Sheet is rapidly thinning, particularly at its two largest outlet glaciers, Pine Island Glacier and Thwaites Glacier, while East Antarctica remains relatively stable. Projections give a mixed picture, some model project mass gain by increased surface mass balance, most models project some or severe mass loss by increasing ice discharge. In this study, we explore the effect of present-day ice thickness change rates on forced future simulations of the Antarctic Ice Sheet using the Community Ice Sheet Model (CISM). We start with a series of schematic, uniform ocean temperature perturbations to probe the sensitivity of the modelled present-day imbalance to ocean warming. We then apply ocean and atmospheric forcing from seven ESMs from the CMIP5 and CMIP6 ensemble to simulate the Antarctic Ice Sheet from 2015 to 2500. The schematic experiments suggest the presence of an ice-dynamical limit, TG cannot collapse before ~2100 without more than 2 degrees of schematic, suddcen and uniform ocean warming. Meanwhile, the maximum GMSL rise rate during the collapse increases linearly with ocean temperature, indicating that while earlier collapse timing shows diminishing returns, the rate of sea-level rise keeps on intensifying with stronger forcing. In the simulations driven with ESM forcing, including or excluding the present-day imbalance contributes for the West Antarctic Ice Sheet as much to the uncertainty in the mass loss rates in the coming 5 centuries as the choice of ESM forcing. For the East Antarctic Ice Sheet on shorter timescales (until 2100), adding the present-day observed mass change rates doubles its global mean sea level rise contribution. On longer timescales (2100–2500), the effect of the present-day observed mass change rates is smaller. Thinning of the West Antarctic Ice Sheet induced by the present-day imbalance is to a small degree partly compensated by present-day ice sheet thickening of the East Antarctic Ice Sheet over the coming centuries, which persists in our simulations. Moreover, these deviations are overshadowed by the mass losses induced by the projected ocean warming. The relative importance of including the observed present-day mass loss rates decreases for larger (ocean) warming under climate forcing, and decreases over time.

Year:

2025

A realistic representation of the Antarctic sea ice surface albedo, especially during the spring and summer periods, is essential to obtain reliable atmospheric and oceanic model predictions. We used regional climate (HCLIM, MAR, RACMO), regional oceanic (MetROMS-UHel, NEMO) models and ERA5 reanalysis to investigate how well these models describe the basic sea ice characteristics: sea ice albedo, snow and ice thickness. We analyse models against a range of observations, including in-situ measurements from the ISPOL (Weddell Sea, Dec. 2004) and Marsden (McMurdo Sound, Nov. 2022) field campaigns, as well as drone and satellite data. Models perform well in reproducing the sea ice in certain conditions: during the ISPOL campaign, characterised by thicker snow cover and mild weather that resulted in daytime melt-driven albedo changes and nighttime refreezing in the snow-covered sea ice most models did well; MetROMS-UHel, NEMO, HCLIM and MAR reproduce mean values found in observations, and MetROMS-UHel captures even the observed diurnal albedo variability. However, all models had difficulty reproducing the sea ice conditions in the McMurdo Sound. The observed mean surface albedo was largely influenced by variations in drifting snow accumulation patterns over very thin (few to few tens of cm) snow cover and most models clearly overestimated the albedo. Over the colder and drier sea-ice regions with thin or patchy snow cover, the key issues affecting the accuracy of albedo models are the treatment of fractional snow cover and the snow albedo dependence on snow depth. Over the broader Weddell and Ross seas, sea ice albedo is primarily determined by sea ice concentration fields. HCLIM, MAR, and RACMO rely on ERA5 input for sea ice concentration fields, whereas MetROMS-UHel and NEMO calculate them internally, resulting in differences in both sea ice concentration and albedo patterns. Albedo parameterisations are still relevant: RACMO and ERA5 predict significantly darker sea ice over the Weddell Sea during the ISPOL campaign, while their predictions align better with observations over the Ross Sea during the Marsden campaign. Sea ice albedo is typically parameterised in models as a function of one or more variables, including air temperature, surface temperature, snow/ice type, snow grain size, snow depth, density, sea ice thickness, cloud cover fraction and solar zenith angle. The simplest approaches, like those in ERA5 and RACMO, rely on prescribed sea ice albedo values based on Ebert and Curry (1993). In HCLIM, the intermediate-complexity snow model determines snow reflectivity based on snow grain size distribution, which is only a function of snow density, and the bare ice albedo follows a simple temperature-based relationship. When more sophisticated radiative transfer schemes are applied, albedo is calculated based on the inherent optical properties of the surface, such as in MetROMS-UHel. Integrating advanced radiative transfer models to the regional climate or ocean models, represents a significant advancement in simulating surface processes.

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2025

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Year:

2024

Collaborators (25)

Carleen Reijmer

Associate Professor

Utrecht University

NETHERLANDS
View details

Kyle Clem

Lecturer in Climate Science

Victoria University of Wellington

NEW ZEALAND
View details

Pranab Deb

Assistant Professor

Indian Institute of Technology, Kharagpur

View details

Xun Zou

Atmospheric Scientist

Scripps Institution of Oceanography

UNITED STATES
View details

Christoph Kittel

University of Liege

BELGIUM
View details

Lorenz Meire

Senior scientist

-

NETHERLANDS
View details

Annette Rinke

Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research

GERMANY
View details

Srinidhi Gadde

Assistant Professor

University of Twente

NETHERLANDS
View details

David Bromwich

Research Professor

Ohio State University

UNITED STATES
View details

Peter Kuipers Munneke

Assistant professor

Utrecht University

NETHERLANDS
View details

Vincent Favier

Physicist (Professor)

Université Grenoble Alpes

FRANCE
View details

X. Fettweis

Full Prof

University of Liege

BELGIUM
View details

Gerd-Jan van Zadelhoff

Koninklijk Nederlands Meteorologisch Instituut

NETHERLANDS
View details

Michiel Roland van den Broeke

Utrecht University

NETHERLANDS
View details

Nicolas Jourdain

Reseacher

CNRS en Alpes

FRANCE
View details

Heiko Goelzer

Research Professor

-

NORWAY
View details

Max Brils

Northumbria University

UNITED KINGDOM
View details

Brice Noël

-

BELGIUM
View details

Walter Immerzeel

Professor

Utrecht University

NETHERLANDS
View details

Maurice Van Tiggelen

Utrecht University

NETHERLANDS
View details

JOHN CASSANO

University of Colorado Boulder

UNITED STATES
View details

Petteri Uotila

University of Helsinki

FINLAND
View details

René Wijngaard

Utrecht University

NETHERLANDS
View details

Ella Gilbert

University of Reading

UNITED KINGDOM
View details

Ruzica Dadic

WSL-Institut für Schnee- und Lawinenforschung SLF

SWITZERLAND
View details