Ice Model Research Articles

The Role of the ICE Model in Doctor-Patient Shared Decision Making

The ICE model plays a crucial role in shared decision-making between patients and healthcare providers. Adequate information helps patients better understand their health conditions and make informed choices; effective communication facilitates building mutual trust and reaching consensus between providers and patients; and the demonstration of empathy can alleviate patients' anxiety and enhance their confidence in participating in decision-making. In recent years, the application of the ICE model in clinical practice has gained increasing attention. This article mainly explores and analyzes the application of the ICE model in medical practice and its impact on the quality of patient-provider interactions. The research aims to evaluate how the ICE model promotes effective bidirectional communication between patients and providers, and how providers' empathy can enhance psychological support for patients and treatment adherence. Through this study, the goal is to comprehensively understand the role of the ICE model in improving the quality of healthcare services, strengthening patient-provider relationships, and assisting providers in delivering more personalized, patient-centered care. Finally, current challenges are discussed, and future development directions are proposed.

On the sensitivity of sea ice deformation statistics to plastic damage

Abstract. We implement a plastic damage parametrization, distinct from the elastic damage in the elasto-brittle framework, in the standard viscous–plastic (VP) sea ice model to disentangle its effect from resolved model physics (visco-plastic with and without damage) on its ability to reproduce observed scaling laws of deformation. To this end, we compare scaling properties and multifractality of simulated divergence and shear strain rate, as proposed in the Sea Ice Rheology Experiment (SIREx) studies, with those derived from the RADARSAT Geophysical Processor System (RGPS). Results show that including a plastic damage parametrization in the standard viscous–plastic model increases the spatial but decreases the temporal localization of simulated linear kinematic features (LKFs) and brings all spatial deformation rate statistics in line with observations from RGPS without the need to increase the mechanical shear strength of sea ice as recently proposed for lower-resolution viscous–plastic sea ice models. In fact, including damage with a healing timescale of th=30 d and an increased mechanical strength unveils multifractal behavior that does not fit the theory. Therefore, a plastic damage parametrization is a powerful tuning knob affecting the deformation statistics of viscous–plastic sea ice.

Numerical Simulation of Water Film Flow and Breakup on Anti-Icing Surface

The flow and morphological characteristics of liquid water on the icing and anti-icing surfaces of aircraft are closely related to the icing characteristics and anti-icing surface temperature distribution. To predict the flow and breakup characteristics of a water film, a 3D model of continuous water film flow and a model of water film breakup into rivulets on an anti-icing surface were constructed based on the icing model, and the corresponding methods for solving the models were developed. Using the NACA0012 airfoil as a simulation object, the changing characteristics of height and velocity for a continuous water film with time and the morphological characteristics of rivulets formed from the breakup of a continuous water film were simulated numerically. The results indicate that, with an increase in inflow velocity, the time required for the water film to completely cover the surface and reach stability decreases. Downstream in the water droplet impact zone, the calculated values of continuous water film height align well with experiments, as well as the stream height at the continuous water film rupture location with the experimental values. With the reasonable contact angle, the calculation error of the stream width is within 10%.

Towards a Model of Snow Accretion for Autonomous Vehicles

Snow accumulation on surfaces exposed to adverse weather conditions has been studied over the years due to a variety of problems observed in different industry sectors, such as aeronautics and wind and civil engineering. With the growing interest in autonomous vehicles (AVs), this concern extends to advanced driver-assistance systems (ADAS). Weather stressors, such as snow and icing, negatively influence the sensor functionality of AVs, and their autonomy is not guaranteed by manufacturers during episodes of intense weather precipitation. As a basis for mitigating the negative effects caused by heavy snowfall, models need to be developed to predict snow accumulation over critical surfaces of AVs. The present work proposes a framework for the study of snow accumulation on road vehicles. Existing icing and snow accretion models are reviewed, and adaptations for automotive applications are discussed. Based on the new capabilities developed by the Weather on Wheels (WoW) program at Ontario Tech University, a model architecture is proposed in order to progress toward adequate snow accretion predictions for autonomous vehicle operating conditions, and preliminary results are presented.

From snow accumulation to snow depth distributions by quantifying meteoric ice fractions in the Weddell Sea

Abstract. Year-round snow cover is a characteristic of the entire Antarctic sea ice cover, which has significant implications for the energy and mass budgets of sea ice, e.g., by preventing surface melt in summer and enhancing sea ice growth through extensive snow ice formation. However, substantial observational gaps in the seasonal cycle of Antarctic sea ice and its snow cover limit the understanding of important processes in the ice-covered Southern Ocean. They also introduce large uncertainties in satellite remote sensing applications and climate studies. Here we present results from 10 years of autonomous snow observations from Snow Buoys in the Weddell Sea. To distinguish between actual snow depth and potential snow ice thickness within the accumulated snowpack, a one-dimensional thermodynamic sea ice model is applied along the drift trajectories of the buoys. The results show that potential snow ice formation, with an average maximum thickness of 35 cm, was detected along 41 % of the total track length of the analyzed Snow Buoy tracks, which corresponds to about one-quarter of the snow accumulation. In addition, we simulate the evolution of internal snow properties along the drift trajectories with the more complex SNOWPACK model, which results in superimposed ice thicknesses between 0 and 14 cm on top of the snow ice layer. These estimates will provide an important reference dataset for both snow depth and meteoric ice rates in the Southern Ocean.

Surface and Atmospheric Heating Responses to Spectrally Resolved Albedo of Frozen and Liquid Water Surfaces

AbstractMultiple Earth system models (ESMs) discretize surface albedo into two semi‐broadbands comprising the UV/visible and near‐infrared (NIR) wavelengths. Here, we use an offline single‐column radiative transfer model to investigate the radiative effects of spectrally resolving the surface albedo. We use the Snow, Ice, and Aerosol Radiative model, extended to simulate liquid water, to calculate snow, ice, and liquid water albedo. We flux‐weight the hyperspectral albedo into the coarser spectral bands used by the atmospheric shortwave radiative transfer model. We establish representative atmospheric profiles for the three surface types and compare their shortwave fluxes and atmospheric warming rates with the spectrally resolved albedo to those calculated with the semi‐broadband approximation. Spectrally resolved surface albedo over snow and ice reduces atmospheric warming by darkening the albedo of NIR bands, correcting the too‐strong surface absorption in visible bands, and too‐weak surface absorption in shortwave infrared bands caused by the semi‐broadband approximation. We explore the effects on surface and atmospheric warming rates of varying solar zenith angle, cloud cover, relative humidity, and snow grain/air bubble radii. The semi‐broadband albedo biases can exceed 10% and 2% for the surface and atmospheric net flux respectively, being particularly strong under conditions which alter the distributions of surface insolation (i.e., cloud cover or increased atmospheric water vapor). These results show that transmitting a higher resolution spectral radiation field between the atmosphere and surface reduces biases in surface absorption and atmospheric heating present in ESMs that currently use the semi‐broadband approximation.

DISCUSSION OF THE MECHANISM OF THE ELECTRICAL CONDUCTIVITY OF PURE WATER IN THE FRAMEWORK OF THE BJERRUM MODEL

An alternative mechanism of electrical conductivity and ionization of molecules in pure water is considered. Calculation of the temperature dependence for the activation energy of ionization of water molecules was carried out using data on the value of its ionic product. The calculated activation energies for the ionization and flow processes show the presence of a certain proportionality between them. This suggests that the ionization of water is caused by intermolecular forces rather than intramolecular forces. Accordingly, the electrical conductivity of water is associated with the appearance and movement in it of L and D defects of hydrogen bonds between water molecules, which have significant electrical pseudo charges and are considered in the Bjerrum model for ice. This work proposes a new interpretation of the mechanism of ionization (pseudoionization) of molecules and generation of L and D defects in water. Rotation is considered not for one, but for two of its molecules in opposite directions with partial compensation of their angular momentum, and the breaking of not three bonds for one molecule, but four bonds with neighbors for two molecules. Moreover, these four neighbors do not include molecules from water clusters that have stronger bonds, since neighboring molecules with weaker bonds can be found for the process of rotation activation. Accordingly, the exponential contribution made by molecules from water clusters is absent or small in the ionization activation energy. In this way, the activation energy for the flow, which occurs throughout the entire volume of the liquid and automatically includes molecules from clusters, differs from the activation energy of electrical conductivity, which occurs for individual molecules in chains of bonds. The presence of a maximum in the ionic product of water near 250 ⁰C can be explained by the rapid decrease in the lifetime of L and D defects with heating, which prevents the use of the standard equation for accurate calculation of the ionization activation energy above 100...200 ⁰C. However, the presence of this maximum contradicts the generally accepted model of proton hopping between neighboring hydronium and water molecules in the mechanism of its electrical conductivity.

Effects of Ice-Microstructure-Based Inherent Optical Properties Parameterization in the CICE Model

The constant inherent optical properties (IOPs) for sea ice currently applied in sea ice models do not realistically represent the dividing of shortwave radiative fluxes in sea ice and the ocean below it. Here we implement a parameterization of variable IOPs based on ice microstructures in the Los Alamos sea ice model, version 6.0 (CICE6) and investigate its effects on the simulation of the dividing of shortwave radiation and sea ice in the Arctic. Our sensitivity experiments indicate that variable IOP parameterization results in strong seasonal variation for the IOP parameters, typically reaching the seasonal maximum in the boreal summer. With such large differences, variable IOP parameterization leads to increased absorbed solar radiation at the surface and in the interior of Arctic sea ice relative to constant IOPs, up to ~3 W/m2, but decreased solar radiation penetrating into the ocean, up to ~5–6 W/m2. The changes in the dividing of shortwave fluxes in sea ice and the ocean below it induced by the variable IOPs have significant influence on Arctic sea ice thickness by modulating surface and bottom melting and frazil ice formation (increasing surface melting by ~16% and reducing bottom melting by ~11% in summer).

An overview of the NEMO modelling for the BaySys project

This article is intended as an introduction to discuss the development of a modelling framework to examine simulated climate change and river discharge regulation and their combined impact on marine conditions in the Hudson Bay Complex as a contribution to BaySys, a collaborative project between Manitoba Hydro, Hydro-Quebec, the University of Manitoba, the University of Alberta, Université Laval and Ouranos. In support of this work, a sea ice and oceanographic model was improved and then used to further study the effects of freshwater loading and ice cover on the circulation of Hudson Bay. This modelling perspective is based on the Nucleus for European Modelling of the Ocean (NEMO) ocean general circulation model coupled to version 2 of the Louvain-la-Neuve sea ice model (LIM2). The goal of the modelling was to provide a framework and tool for simulating projected changes in marine state and dynamic variables, while also enabling an integration of observations and numerical analyses. A key aspect of this work was the climate-hydrologic-ocean model integration aspect. The inclusion of a biogeochemical model and explicit tidal forcing to examine the evolution of a Canadian marginal sea with century-long integrations was also a novel aspect of the work. Overall, this work examines the NEMO modelling configuration used in BaySys, how it is set up and the experiments carried out. A broader picture evaluation of the model output is made including the BaySys mooring observations, showing that the modelling framework is suitable to examine the posed questions on the role of climate change and river regulation.

The U-net-based ice pore parameter extraction method for establishing the SAW icing sensing mechanism

The porous ice layer formed upon the SAW (surface acoustic wave) propagation path will produce transient acoustic attenuation, and a new icing sensing method can be constructed based on this feature. Obtaining the pore characteristics of an ice layer is a prerequisite and basis for deriving the sensing mechanism of SAW icing. Employing the U-net network, a segmentation model of frozen pores is developed by collecting frozen images and calibrating sample data sets for training in this work. The MIOU and the MPA of the model reach 92.16 % and 95.15 % respectively. Then, the watershed algorithm and post-processing were used to accurately obtain the pore characteristic parameters (porosity, pore number, and average pore area) of the ice layer. On this basis, applying the Biot's theory for two-phase porous media, a theoretical model of the SAW icing sensing mechanism was established, and the acoustic sensing response under different icing types was theoretically simulated. To prove the accuracy of the theoretical simulation, a SAW sensor using a waveguide structure of SU-8/ST-90 °X quartz operating at 80 MHz was constructed and evaluated experimentally. The results demonstrate that the acoustic sensing responses of different ice types are completely consistent with the theoretical results. This fully verifies the accuracy of the ice pore parameters extracted in this work and also confirms the feasibility of using pore microscopic information to achieve ice detection and ice type discrimination.

Modeling and analysis for volatile characteristics of lunar water ice

Lunar water ice resources are important space assets for human survival and sustainable living on the Moon. Detection data indicates that the permanently shadowed regions at the lunar poles contain water ice resources. The efficiency of in-situ extraction of lunar water ice resources depends on its volatile characteristics. In the context of extraction and utilization of water ice resources in the permanently shadowed regions, the sublimation and diffusion processes of lunar water ice are investigated by combining the theory of pure ice sublimation with the fractal theory of porous media. On this basis, a volatilization model for lunar water ice is established to examine the effect of temperature, relative density, and water content on the volatile behaviors of lunar water ice. The findings of this study can provide technical and theoretical support for the in-situ exploration, extraction, and utilization of lunar water ice resources and offer insights into the existence and distribution of lunar water ice resources.

Modeling the Winter Heat Conduction Through the Sea Ice System During MOSAiC

AbstractModels struggle to accurately simulate observed sea ice thickness changes, which could be partially due to inadequate representation of thermodynamic processes. We analyzed co‐located winter observations of the Arctic sea ice from the Multidisciplinary Drifting Observatory for the Study of the Arctic Climate for evaluating and improving thermodynamic processes in sea ice models, aiming to enable more accurate predictions of the warming climate system. We model the sea ice and snow heat conduction for observed transects forced by realistic boundary conditions to understand the impact of the non‐resolved meter‐scale snow and sea ice thickness heterogeneity on horizontal heat conduction. Neglecting horizontal processes causes underestimating the conductive heat flux of 10% or more. Furthermore, comparing model results to independent temperature observations reveals a ∼5 K surface temperature overestimation over ice thinner than 1 m, attributed to shortcomings in parameterizing surface turbulent and radiative fluxes rather than the conduction. Assessing the model deficiencies and parameterizing these unresolved processes is required for improved sea ice representation.

Sea Ice Loss, Water Vapor Increases, and Their Interactions with Atmospheric Energy Transport in Driving Seasonal Polar Amplification

Abstract The ice–albedo feedback associated with sea ice loss contributes to polar amplification, while the water vapor feedback contributes to tropical amplification of surface warming. However, these feedbacks are not independent of atmospheric energy transport, raising the possibility of complex interactions that may obscure the drivers of polar amplification, in particular its manifestation across the seasonal cycle. Here, we apply a radiative transfer hierarchy to an idealized aquaplanet global climate model coupled to a thermodynamic sea ice model. The climate responses and radiative feedbacks are decomposed into the contributions from sea ice loss, including both retreat and thinning, and the radiative effect of water vapor changes. We find that summer sea ice retreat causes winter polar amplification through ocean heat uptake and release, and the resulting decrease in dry energy transport weakens the magnitude of warming. Moreover, sea ice thinning is found to suppress summer warming and enhance winter warming, additionally contributing to winter amplification. The water vapor radiative effect produces seasonally symmetric polar warming via offsetting effects: enhanced moisture in the summer hemisphere induces the summer water vapor feedback and simultaneously strengthens the winter latent energy transport in the winter hemisphere by increasing the meridional moisture gradient. These results reveal the importance of changes in atmospheric energy transport induced by sea ice retreat and increased water vapor to seasonal polar amplification, elucidating the interactions among these physical processes.

The Effect of Physically Based Ice Radiative Processes on Greenland Ice Sheet Albedo and Surface Mass Balance in E3SM

AbstractA significant portion of surface melt on the Greenland Ice Sheet (GrIS) is due to dark ice regions in the ablation zone, where solar absorption is influenced by the physical properties of the ice, light absorbing constituents (LACs), and the overlying crustal surface or melt ponds. Earth system models (ESMs) typically prescribe the albedo of ice surfaces as a constant value in the visible and near‐infrared spectral regions. This work advances ESM ice radiative transfer modeling by (a) incorporating a physically based radiative transfer model (SNow, ICe and Aerosol Radiation model Adding‐Doubling Version 4; SNICAR‐ADv4) into the Energy Exascale Earth System Model (E3SM), (b) determining spatially and temporally varying bare ice physical properties over the GrIS ablation zone from satellite observations to inform SNICAR‐ADv4, and (c) assessing the impacts on simulated GrIS albedo and surface mass balance associated with modeling of more realistic bare ice albedo. GrIS‐wide bare ice albedo in E3SMv2 is overestimated by ∼4% in the visible and ∼7% in the near‐infrared wavelengths compared to the Moderate Resolution Imaging Spectroradiometer. Our bare ice physical property retrieval method found that LACs, ice crustal surfaces, and melt ponds reduce visible albedo by 30% in the bare ice region of the GrIS ablation zone. The realistic bare ice albedo reduces surface mass balance by ∼145 Gt, or 0.4 mm of sea‐level equivalent between 2000 and 2021 compared to the default E3SM. This work highlights the importance of simulating bare ice albedo accurately and realistically to improve our ability to quantify changes in the GrIS surface mass and radiative energy budgets.

Subsurface Warming of the West Antarctic Continental Shelf Linked to El Niño‐Southern Oscillation

AbstractRecent observations suggest that El Niño–Southern Oscillation (ENSO) impacts basal melting of West Antarctic ice shelves, yet sparse ocean observations limit our understanding of the associated processes. Here we investigate how ENSO events modulate subsurface West Antarctic shelf temperatures using high‐resolution global ocean‐sea ice model simulations. During El Niño, the subsurface shelf warming between 150 m and the shelf bottom can be up to 0.5°C in front of ice shelves. This warming arises from a weaker Amundsen Sea Low (ASL) and weaker coastal easterlies that reduce on‐shelf Ekman transport of cold surface waters, enabling enhanced transport of warm Circumpolar Deep Water (CDW) onto the shelf. A largely opposite response occurs during La Niña, with a stronger ASL and stronger Ekman transport that results in less cross‐shelf CDW transport and cooling in the subsurface. These findings have implications for interpreting basal melting on interannual to decadal time‐scales in West Antarctica.

Assimilation of satellite swaths versus daily means of sea ice concentration in a regional coupled ocean–sea ice model

Abstract. Operational forecasting systems routinely assimilate daily means of sea ice concentration (SIC) from microwave radiometers in order to improve the accuracy of the forecasts. However, the temporal and spatial averaging of the individual satellite swaths into daily means of SIC entails two main drawbacks: (i) the spatial resolution of the original product is blurred (especially critical in periods with strong sub-daily sea ice movement), and (ii) the sub-daily frequency of passive microwave observations in the Arctic are not used, providing less temporal resolution in the data assimilation (DA) analysis and, therefore, in the forecast. Within the SIRANO (Sea Ice Retrievals and data Assimilation in NOrway) project, we investigate how challenges (i) and (ii) can be avoided by assimilating individual satellite swaths (level 3 uncollated) instead of daily means (level 3) of SIC. To do so, we use a regional configuration of the Barents Sea (2.5 km grid) based on the Regional Ocean Modeling System (ROMS) and the Los Alamos Sea Ice Model (CICE) together with the ensemble Kalman filter (EnKF) as the DA system. The assimilation of individual swaths significantly improves the EnKF analysis of SIC compared to the assimilation of daily means; the mean absolute difference (MAD) shows a 10 % improvement at the end of the assimilation period and a 7 % improvement at the end of the 7 d forecast period. This improvement is caused by better exploitation of the information provided by the SIC swath data, in terms of both spatial and temporal variance, compared to the case when the swaths are combined to form a daily mean before assimilation.

Role of Ice Mechanics on Snow Viscoplasticity

AbstractThe porous structure of snow becomes denser with time under gravity, primarily due to the creep of its ice matrix with viscoplasticity. Despite investigation of this behavior at the macroscopic scale, the driving microscopic mechanisms are still not well understood. Thanks to high‐performance computing and dedicated solvers, we modeled snow elasto‐viscoplasticity with 3D images of its microstructure and different mechanical models of ice. The comparison of our numerical experiments to oedometric compression tests measured by tomography showed that ice in snow rather behaves as a heterogeneous set of ice crystals than as homogeneous polycrystalline ice. Similarly to dense ice, the basal slip system contributed at most, in the simulations, to the total snow deformation. However, in the model, the deformation accommodation between crystals was permitted by the pore space and did not require any prismatic and pyramidal slips, whereas the latter are pre‐requisite for the simulation of dense ice.

Microscopic mechanism of plastic heterogeneous deformation of columnar-grained polycrystalline ice

Studying ice deformation starting from crystal scale can provide insight into the microscopic mechanisms of ice deformation. In this paper, a crystal plasticity constitutive model of ice considering the dislocation slip mechanism and grain-size dependence was implemented within the finite element framework, and a modeling approach starting from ice single crystals based on the crystal plasticity finite element method (CPFEM) was proposed to study the elastic-viscoplastic deformation of polycrystalline ice samples. Finite element models of ice single crystals and columnar-grained polycrystalline ice characterizing the microstructure were established successively, and the proposed method was validated by comparing the uniaxial compression prediction results of ice single crystals and columnar-grained polycrystalline ice with the existing experimental data. Finally, the microscopic mechanism of plastic heterogeneous deformation of columnar-grained polycrystalline ice under constant strain rate compression conditions was analyzed. The results indicate that the modeling approach from ice single crystals to polycrystalline ice samples based on CPFEM can reasonably describe the mechanical properties of ice deformation. The deformation of columnar-grained polycrystalline ice globally exhibits apparent slip bands around 30° - 60° relative to the loading direction and locally presents intragranular localization bands parallel or perpendicular to the C-axis. This plastic heterogeneous deformation phenomenon is related to the elastic-viscoplastic anisotropy of ice single crystals and the interaction of neighboring grains at grain boundaries (GBs).

The linkage between wintertime sea ice drift and atmospheric circulation in an Arctic ice-ocean coupled simulation

By analyzing an Arctic ice-ocean coupled simulation, we study the linkage between wintertime sea ice drift and atmospheric circulation, and interpret the driving force terms in the sea ice dynamic equation. Sea ice drift anomaly is featured by an anticyclonic (cyclonic) gyre when regulated by negative (positive) phase of Arctic Oscillation with positive (negative) phase of Arctic Dipole, and a quasi-meridional stream from Chukchi-Beaufort (Barents-Kara) Seas to Barents-Kara (Chukchi-Beaufort) Seas when regulated by positive (negative) phase of Arctic Oscillation with positive (negative) phase of Arctic Dipole. Sea ice drift anomaly, when regulated by the mode alone, resembles spatial pattern of leading atmospheric mode. Decomposing sea ice dynamical equation shows that wind-ice stress dominates sea ice drift in areas away from islands and continental coastlines, ocean-ice stress acts as a resistant power to partly cancel the wind-ice stress in these areas, while in the coastal areas such as the thick multiyear ice zone north of the Canadian Arctic Archipelago, the wind-ice and ocean-ice stresses are small, the balance exists between sea surface height potential gradient and internal ice stress divergence. Developing more sophisticated internal ice stress expression in ice model is of great important to correctly project future sea ice change for the ice modeling community.

Parameter sensitivity analysis of a sea ice melt pond parametrisation and its emulation using neural networks

Accurate simulation of sea ice is critical for predictions of future Arctic sea ice loss, looming climate change impacts, and more. A key feature in Arctic sea ice is the formation of melt ponds. Each year melt ponds develop on the surface of the ice and primarily via affecting the albedo, they have an enormous effect on the energy budget and climate of the Arctic. As melt ponds are subgrid scale and their evolution occurs due to a number of competing, poorly understood factors, their representation in models is parametrised.Sobol sensitivity analysis, a form of variance based global sensitivity analysis is performed on an advanced melt pond parametrisation (MPP), in Icepack, a state-of-the-art thermodynamic column sea ice model. Results show that the model is very sensitive to changing its uncertain MPP parameter values, and that these have varying influences over model predictions both spatially and temporally. Such extreme sensitivity to parameters makes MPPs a potential source of prediction error in sea-ice models, given that the (often many) parameters in MPPs are usually poorly known.Machine learning (ML) techniques have shown great potential in learning and replacing subgrid scale processes in models. Given the complexity of melt pond physics and the need for accurate parameter values in MPPs, we propose an alternative data-driven MPPs that would prioritise the accuracy of albedo predictions. In particular, we constructed MPPs based either on linear regression or on nonlinear neural networks, and investigate if they could substitute the original physics-based MPP in Icepack.Our results shown that linear regression are insufficient as emulators, whilst neural networks can learn and emulate the MPP in Icepack very reliably. Icepack with the MPPs based on neural networks only slightly deviates from the original Icepack and overall offers the same long term model behaviour.We then searched for the smallest possible emulator that achieves good performance by performing features selection based on mutual information. Results indicates that a smaller model, based only on a portion of the full set of input variables needed by the physical MPP, is also sufficient to approximate and replace the physical MPP. This smaller emulators are not only computationally faster but also easier to interpret on a physical ground.Several and diverse challenges still exist, yet this study is an encouraging first step, prior to using real data, towards the adoption of data-driven MPPs in sea-ice models.