Search

Modeling and understanding sea ice dynamics in marginal ice zones rely on measurements of sea ice. Lagrangian observations of ice floes provide insight into the dynamics of sea ice, the ocean, and the atmosphere. However, optical satellite images are susceptible to atmospheric noise, leading to gaps in the retrieved time series of floe positions. This paper presents an efficient and statistically accurate nonlinear dynamical interpolation framework for recovering missing floe observations. It exploits a balanced physics‐based and data‐driven construction to address the challenges posed by the high‐dimensional and nonlinear nature of the coupled atmosphere‐ice‐ocean system, where effective reduced‐order stochastic models, nonlinear data assimilation, and simultaneous parameter estimation are systematically integrated. The new method succeeds in recovering the locations, curvatures, angular displacements, and the associated strong non‐Gaussian distributions of the missing floes in the Beaufort Sea. It also accurately estimates floe thickness and recovers the unobserved underlying ocean field with an appropriate uncertainty quantification, advancing our understanding of Arctic climate. Plain Language Summary: Tracking individual ice floes is a unique measurement of areas of the Arctic where the ice cover interacts with the open ocean. Unfortunately, optical satellite images of these areas are frequently obscured by clouds, leading to missing observations of the ice floes. Traditional methods of filling in these gaps in the data set have issues. Linear interpolation, which averages between available observations to fill in missing ones, fails to recover the curvature of the floes. Dynamical interpolation methods, which take into account the physical properties of the ice floes, are very computationally expensive. This paper presents a nonlinear dynamical interpolation framework for recovering missing floe observations, which is both computationally efficient and statistically accurate. The framework incorporates a model of the atmosphere, ocean, and sea ice and systematically develops data‐driven reduced‐order stochastic models, which significantly accelerate the dynamical interpolation while retaining accuracy. In addition, the framework estimates key physical parameters, such the floe thickness. This new method succeeds in recovering the locations, curvatures, angular displacements, and strong non‐Gaussian statistics of the missing floes in a data set of ice floes in the Beaufort Sea. These results can provide complete data sets that advance our understanding of Arctic climate. Key Points: Developed systematic data‐driven reduced‐order stochastic model to significantly lower the computational cost for dynamical interpolationEstimated key physical parameters such as the floe thickness and recovered the unobserved ocean flow field with uncertainty quantificationRecovered important non‐Gaussian distributions of the curvature and angular displacement of the missing floes in the Beaufort Sea [ABSTRACT FROM AUTHOR]

To assess the effect of ocean‐atmosphere coupling in the climate response to forced sea ice loss, the Polar Amplification Model Intercomparison Project protocol includes centennial coupled atmosphere‐ocean general circulation model simulations with imposed sea ice loss. The protocol, which specifies sea ice concentration and thickness distribution targets, does not prescribe a method for achieving them. Although different methods for imposing sea ice loss (or growth) in models have been documented, testing of the method‐dependence of the resulting climate responses has been limited. Achieving the targeted sea ice state has proven to be challenging using the so‐called ghost‐flux nudging method, which induces ice melt from below, as this method does not constrain the partitioning between thickness and concentration. We propose, describe and test a simple method that combines the advantages of direct sea ice nudging and ghost‐flux nudging. The hybrid nudging method better captures the partitioning between thickness and concentration while conserving total water content. We document how this novel sea ice constraining method reaches specific targets, enhances surface turbulent heat flux responses to sea ice loss, and induces tropospheric warming for both polar regions. Plain Language Summary: The Arctic is warming faster than the global average due to several processes that, once combined, lead to so‐called Arctic Amplification. Part of this anomalous polar warming comes from an intense reduction in ice cover allowing heat into the ocean, warming the Arctic ocean near the surface, hence melting more ice. A joint effort by several climate modeling groups called the Polar Amplification Model Intercomparison Project (PAMIP) aims at better understanding Arctic Amplification through a coordinated set of climate simulations. Among this ensemble of simulations is a set of centennial simulations performed with fully coupled state‐of‐the‐art climate models. In these experiments, Arctic (and Antarctic) sea ice are forced to reach specific states in order to better isolate Arctic Amplification and sea ice loss from the rest of anthropogenic global warming. In this paper, we propose a simple technique to nudge sea ice models to specific states such as prescribed by PAMIP. This new method combines advantages from existing techniques to improve the control over the extent and the thickness of the ice. We document how our novel method produces atmospheric warming and circulation responses that are more consistent with previously published results than the existing techniques. Key Points: Already established sea ice nudging method leads to issues in the partitioning between sea ice concentration (SIC) and sea ice thickness (SIT)New hybrid nudging method significantly improves capture of both SIC and SIT targets for Polar Amplification Model Intercomparison Project, in particular for Arctic sea iceHybrid scheme increases Arctic turbulent heat flux and warming [ABSTRACT FROM AUTHOR]

The discrete element method (DEM) can provide detailed descriptions of sea ice dynamics that explicitly model floes and discontinuities in the ice, which can be challenging to represent accurately with current models. However, floe-scale stresses that inform lead formation in sea ice are difficult to calculate in current DEM implementations. In this paper, we use the ParticLS software library to develop a DEM that models the sea ice as a collection of discrete rigid particles that are initially bonded together using a cohesive beam model that approximates the response of an Euler-Bernoulli beam located between particle centroids. Ice fracture and lead formation are determined based on the value of a non-local Cauchy stress state around each particle and a Mohr-Coulomb fracture model. Therefore, large ice floes are modeled as continuous objects made up of many bonded particles that can interact with each other, deform, and fracture. We generate particle configurations by discretizing the ice in MODIS satellite imagery into polygonal floes that fill the observed ice shape and extent. The model is tested on ice advecting through an idealized channel and through Nares Strait. The results indicate that the bonded DEM model is capable of qualitatively capturing the dynamic sea ice patterns through constrictions such as ice bridges, arch kinematic features, and lead formation. In addition, we apply spatial and temporal scaling analyses to illustrate the model's ability to capture heterogeneity and intermittency in the simulated ice deformation. Plain Language Summary Numerical models of sea ice give researchers important tools to study how the Arctic is changing. Discrete element method (DEM) models idealize sea ice as a collection of individual rigid bodies, or "particles," that can interact with each other independently, and can capture the discontinuities and geometric force concentrations in ice that are common at small scales. In this paper, we extend recent DEM models and evaluate a non-local stress state within the modeled ice (bonded DEM particles) to determine when the ice should fracture. As a result, the model simulates large pieces of ice that can break into smaller pieces, or floes, composed of many still-bonded particles. This allows us to represent both discrete fractures, and emergent aggregate behavior of ice as it deforms. As an example, we simulate ice advecting through Nares Strait. [ABSTRACT FROM AUTHOR]

A new version of the AWI Coupled Prediction System is developed based on the Alfred Wegener Institute Climate Model v3.0. Both the ocean and the atmosphere models are upgraded or replaced, reducing the computation time by a factor of 5 at a given resolution. This allowed us to increase the ensemble size from 12 to 30, maintaining a similar resolution in both model components. The online coupled data assimilation scheme now additionally utilizes sea‐surface salinity and sea‐level anomaly as well as temperature and salinity profile observations. Results from the data assimilation demonstrate that the sea‐ice and ocean states are reasonably constrained. In particular, the temperature and salinity profile assimilation has mitigated systematic errors in the deeper ocean, although issues remain over polar regions where strong atmosphere‐ocean‐ice interaction occurs. One‐year‐long sea‐ice forecasts initialized on 1 January, 1 April, 1 July and 1 October from 2003 to 2019 are described. To correct systematic forecast errors, sea‐ice concentration from 2011 to 2019 is calibrated by trend‐adjusted quantile mapping using the preceding forecasts from 2003 to 2010. The sea‐ice edge raw forecast skill is within the range of operational global subseasonal‐to‐seasonal forecast systems, outperforming a climatological benchmark for about 2 weeks in the Arctic and about 3 weeks in the Antarctic. The calibration is much more effective in the Arctic: Calibrated sea‐ice edge forecasts outperform climatology for about 45 days in the Arctic but only 27 days in the Antarctic. Both the raw and the calibrated forecast skill exhibit strong seasonal variations. Plain Language Summary: Ocean data sparseness and systematic model errors pose problems for the initialization of coupled seasonal forecasts, especially in polar regions. Our global forecast system follows a seamless approach with refined ocean resolution in the Arctic. The new version presented here features higher computational efficiency and utilizes more ocean and sea‐ice observations. Ice‐edge forecasts outperform a climatological benchmark for about 1 month, comparable to established systems. Key Points: We describe an upgrade of the AWI Coupled Prediction System with new ocean and atmosphere models and more observations assimilatedIndependent evaluations show advances in the new version on the analysis of the sea‐ice and ocean states against the old oneCalibrated sea‐ice edge forecasts outperform a climatological benchmark for around 1 month in both hemispheres [ABSTRACT FROM AUTHOR]

Cloud‐phase partitioning has been studied in the context of cloud feedback and climate sensitivity; however, precipitation‐phase partitioning also has a significant role in controlling the energy budget and sea ice extent. Although some global models have introduced a more sophisticated precipitation parameterization to reproduce realistic cloud and precipitation processes, the effects on the process representation of mixed‐ and ice‐phase precipitation are poorly understood. Here, we evaluate how different precipitation modeling (i.e., diagnostic [DIAG] vs. prognostic [PROG] schemes) affects the simulated precipitation phase and occurrence frequency. Two versions of MIROC6 were used with the satellite simulator COSP2. Although the PROG scheme significantly improves the simulated cloud amount and snowfall rates, the phase partitioning, frequency, and intensity of precipitation with the PROG scheme are still biased, and are even worse than with the DIAG scheme. We found a "too frequent and too light" Arctic snowfall bias in the PROG, which cannot be eliminated by model tuning. The cloud‐phase partitioning is also affected by the different approaches used to consider precipitation. The ratio of supercooled liquid water is underrepresented by switching from the DIAG to PROG scheme, because some snowflakes are regarded to be cloud ice. Given that the PROG precipitation retains more snow in the atmosphere, the underestimation becomes apparent when other models incorporate the PROG scheme. This depends on how much precipitation is within the clouds in the model. Our findings emphasize the importance of correctly reproducing the phase partitioning of cloud and precipitation, which ultimately affects the simulated climate sensitivity. Plain Language Summary: This study examined cloud and precipitation phase partitioning (i.e., the ratio between liquid and ice) in the Arctic using the MIROC6 global climate model (GCM). Despite recent advances in precipitation modeling by GCMs, the associations between the macrostructures (i.e., cloud coverage and precipitation rate) and phase partitioning have been little studied. Prognostic treatment of precipitation, which is a more sophisticated parameterization, yields seasonal and annual cloud cover and snowfall that are in better agreement with satellite observations. However, it tends to generate snowfall too frequently and too lightly, resulting in the misrepresentation of precipitation phase partitioning. In addition, there is a risk of overestimating the ratio of cloud ice to cloud liquid by including prognostic precipitation. The bias is difficult to remove by model tuning alone. If the models misrepresent the precipitation phase partitioning, then the bias will further influence feedback processes in a future warming scenario through the snow‐to‐rain phase change, similar to the cloud phase feedback. Our findings emphasize the importance of conducting process‐oriented model evaluations on a regional scale. Key Points: We examined seasonal model biases in precipitation types of Arctic clouds using a satellite simulatorThere are compensating errors between the cloud amount and snowfall, even for the prognostic precipitation schemeRealistic cloud and precipitation processes and their phase partitioning cannot be achieved by model tuning [ABSTRACT FROM AUTHOR]

Correctly representing the snow on sea‐ice has great potential to improve cryosphere‐atmosphere coupling in forecasting and monitoring (e.g., reanalysis) applications, via improved modeling of surface temperature, albedo and emissivity. This can also enhance the all‐weather all‐surface coupled data assimilation for atmospheric satellite radiances. Using wintertime observations from two Arctic field campaigns, SHEBA and N‐ICE2015, and satellite data, we explore the merits of different approaches to represent the snow over sea‐ice in a set of 5‐day coupled forecasts. Results show that representing the snow insulation effects is essential for capturing the wintertime surface temperature variability over sea‐ice and its response to changes in the atmospheric forcing. Modeling the snow over sea‐ice improves the representation of strong cooling events, reduces surface temperature biases in clear‐sky conditions and improves the simulation of surface‐based temperature inversions. In clear‐sky conditions, when using a multi‐layer snow scheme the root‐mean‐squared error in the surface temperature is reduced by about 60% for both N‐ICE2015 and SHEBA. This study also highlights the role of compensating errors in different components of the surface energy budget in the Arctic boundary layer. During warm air intrusions, errors in the surface temperature increase when cloud phase and cloud radiative processes are misrepresented in the model, inducing large errors in the net radiative energy at the surface. This work indicates that numerical weather prediction systems can fully benefit from a better representation of snow over sea‐ice, for example, with multi‐layer snow schemes, combined with improvements to other boundary layer processes including mixed phase clouds. Plain Language Summary: Correctly representing the snow on sea‐ice in coupled numerical weather prediction models has great potential to improve weather forecast and climate monitoring applications, such as climate reanalyses, which are usually produced using such systems. In this study two different methodologies to account for the effect of the snowpack accumulating over the sea‐ice in 5‐day global forecasts are compared. All experiments are performed with the ECMWF Integrated Forecast System. The presence of snow over sea‐ice enables an improved representation of extreme low temperatures occurring in the Arctic, and so reducing surface temperature errors in clear‐sky conditions. This study also investigates feedbacks between errors in processes controlling the surface temperature evolution in the wintertime Arctic boundary layer, and how more realistic representations of the snowpack accumulating on sea‐ice are affected by these. This work indicates that numerical weather prediction systems can fully benefit from a better representation of snow over sea‐ice, for example, with a multi‐layer snow scheme, combined with further improvements to other boundary layer processes. Key Points: Accounting for the snow over sea‐ice enables strong cooling events and extreme low temperatures in the Arctic to be better representedAccounting for the snow over sea‐ice improves the representation of Arctic winter states in the ECMWF Integrated Forecasting SystemCompensating errors between cloud and surface processes are a key factor affecting the near‐surface forecast biases in the Arctic [ABSTRACT FROM AUTHOR]

Basal melting of marine‐terminating glaciers, through its impact on the forces that control the flow of the glaciers, is one of the major factors determining sea level rise in a world of global warming. Detailed quantitative understanding of dynamic and thermodynamic processes in melt‐water plumes underneath the ice‐ocean interface is essential for calculating the subglacial melt rate. The aim of this study is therefore to develop a numerical model of high spatial and process resolution to consistently reproduce the transports of heat and salt from the ambient water across the plume into the glacial ice. Based on boundary layer relations for momentum and tracers, stationary analytical solutions for the vertical structure of subglacial non‐rotational plumes are derived, including entrainment at the plume base. These solutions are used to develop and test convergent numerical formulations for the momentum and tracer fluxes across the ice‐ocean interface. After implementation of these formulations into a water‐column model coupled to a second‐moment turbulence closure model, simulations of a transient rotational subglacial plume are performed. The simulated entrainment rate of ambient water entering the plume at its base is compared to existing entrainment parameterizations based on bulk properties of the plume. A sensitivity study with variations of interfacial slope, interfacial roughness and ambient water temperature reveals substantial performance differences between these bulk formulations. An existing entrainment parameterization based on the Froude number and the Ekman number proves to have the highest predictive skill. Recalibration to subglacial plumes using a variable drag coefficient further improves its performance. Plain Language Summary: In a world of global warming, the melting of glaciers terminating as floating ice tongues into the oceans of Arctic and Antarctic regions allows those glaciers to flow faster and hence to make a considerable contribution to global mean sea‐level rise. Underneath the ice‐ocean interface, turbulent currents of the order of 10 m thickness (so‐called plumes) develop that transport the melt water from the grounding line where the glacier enters the ocean toward the calving front that marks the seaward end of the glacier. At its base, ambient relatively warm and salty ocean water is mixed into the plumes and is vertically transported toward the ice‐ocean interface, where the melting is increased due to the additional heat supply. Understanding these processes is essential for their incorporation into computer models for the prediction of such melt processes. In this study, an accurate simulation model for the water column is constructed that is able to consistently reproduce these processes. The algorithms developed here are proven to provide reliable results also for models with only a few grid points across the plume and can therefore be implemented into climate models with surface‐following coordinates to more accurately simulate future scenarios of sea level rise. Key Points: A vertically resolving model with second‐moment turbulence closure has been constructed for subglacial plumesConvergent numerical formulations for the ocean‐to‐ice fluxes of momentum, freshwater and heat have been derived from an analytical modelModel results are consistent with bulk parameterizations for the entrainment of ambient water [ABSTRACT FROM AUTHOR]

Observations in polar regions show that sea ice deformations are often narrow linear features. These long bands of deformations are referred to as Linear Kinematic Features (LKFs). Viscous‐plastic sea ice models have the capability to simulate LKFs and more generally sea ice deformations. Moreover, viscous‐plastic models simulate a larger number and more refined LKFs as the spatial resolution is increased. Besides grid spacing, other aspects of a numerical implementation, such as the placement of velocities and the associated degrees of freedom, may impact the formation of simulated LKFs. To explore these effects this study compares numerical solutions of sea ice models with different velocity staggering in a benchmark problem. Discretizations based on A‐,B‐, and C‐grid systems on quadrilateral meshes have similar resolution properties as an approximation with an A‐grid staggering on triangular grids (with the same total number of vertices). CD‐grid approximations with a given grid spacing have properties, specifically the number and length of simulated LKFs, that are qualitatively similar to approximations on conventional Arakawa A‐grid, B‐grid, and C‐grid approaches with half the grid spacing or less, making the CD‐discretization more efficient with respect to grid resolution. One reason for this behavior is the fact that the CD‐grid approach has a higher number of degrees of freedom to discretize the velocity field. The higher effective resolution of the CD‐discretization makes it an attractive alternative to conventional discretizations. Plain Language Summary: Sea ice in the Arctic and Antarctic Oceans plays an important role in the exchange of heat and freshwater between the atmosphere and the ocean and hence in the climate in general. Satellite observations of polar regions show that the ice drift sometimes produces long features that are either cracks (leads) and zones of thicker sea ice (pressure ridges). This phenomenon is called deformation. It is mathematically described by the non‐uniform way in which the ice moves. For numerical models of sea ice motion it is difficult to represent this deformation accurately. Details of the numerics may affect the way these models simulate leads and ridges, their number and length. Specifically, we find by comparing different numerical models, that the way the model variables are ordered on a computational grid to solve the mathematical equations of sea ice motion has an effect of how many deformation features can be represented on a grid with a given spacing between grid points. A new discretization (ordering of model variables) turns out to resolve more details of the approximated field than traditional methods. Key Points: The placement of the sea ice velocity has a mayor influence on the number of simulated linear kinematic features (LKFs)The CD‐grid resolves twice as many LKFs compared to A, B, C‐gridsA, B, C‐grids on quadrilateral meshes resolve a similar number of LKFs as A‐grids on triangular meshes (with the same total number of nodes) [ABSTRACT FROM AUTHOR]

We have equipped the unstructured‐mesh global sea‐ice and ocean model FESOM2 with a set of physical parameterizations derived from the single‐column sea‐ice model Icepack. The update has substantially broadened the range of physical processes that can be represented by the model. The new features are directly implemented on the unstructured FESOM2 mesh, and thereby benefit from the flexibility that comes with it in terms of spatial resolution. A subset of the parameter space of three model configurations, with increasing complexity, has been calibrated with an iterative Green's function optimization method to test the impact of the model update on the sea‐ice representation. Furthermore, to explore the sensitivity of the results to different atmospheric forcings, each model configuration was calibrated separately for the NCEP‐CFSR/CFSv2 and ERA5 forcings. The results suggest that a complex model formulation leads to a better agreement between modeled and the observed sea‐ice concentration and snow thickness, while differences are smaller for sea‐ice thickness and drift speed. However, the choice of the atmospheric forcing also impacts the agreement of the FESOM2 simulations and observations, with NCEP‐CFSR/CFSv2 being particularly beneficial for the simulated sea‐ice concentration and ERA5 for sea‐ice drift speed. In this respect, our results indicate that parameter calibration can better compensate for differences among atmospheric forcings in a simpler model (i.e., sea‐ice has no heat capacity) than in more realistic formulations with a prognostic sea‐ice thickness distribution and sea ice enthalpy. Plain Language Summary: The role of model complexity in determining the performance of sea‐ice numerical simulations is still not completely understood. Some studies suggest that a more sophisticated description of the sea‐ice physics leads to simulations that agree better with sea‐ice observations. Others, however, fail to establish a link between complex model formulations and improved model performance. Here, we investigate this open question by analyzing a set of sea‐ice simulations performed with a revised and improved sea‐ice model that features substantial modularity in terms of model complexity. Ten model parameters in three different model configurations are optimized to improve the agreement between model results and observations, allowing a fair comparison between model configurations with varying complexity. The model optimization is repeated for two different atmospheric forcings to shed light on the relationship between model complexity and other sources of uncertainty in the sea‐ice simulations, such as those associated with the atmospheric conditions. The results suggest that a more complex formulation of our model can lead to a more appropriate representation of sea ice concentration and snow thickness, while it is less relevant for sea‐ice thickness and drift. Key Points: Increased sea‐ice model complexity can improve the simulated sea‐ice concentration and snow thicknessSea‐ice thickness and drift are only weakly affected by model complexityParameter calibration can better compensate for differences between atmospheric forcings in a simpler model [ABSTRACT FROM AUTHOR]

The IPSL‐CM6‐LR atmosphere‐ocean coupled model exhibits a pronounced multicentennial variability of the Atlantic meridional overturning circulation (AMOC), driven by delayed freshwater accumulation and release in the Arctic. The AMOC fluctuations are preceded by salinity‐driven density anomalies in the main deep convection sites in the Labrador and Greenland seas. During a strong AMOC, a combination of reduced sea ice volume and anomalous currents reduces the freshwater export from the Arctic and leads instead to a slow accumulation of freshwater in the central Arctic. Simultaneously, the saltier Atlantic inflow through the Barents Sea results in a positive salinity anomaly in the Eastern Arctic subsurface. When the surface Central Arctic freshwater pool finally reaches the Lincoln Sea, the oceanic currents around Greenland reorganize, leading to the export of the anomalous Arctic freshwater to the North Atlantic, enhancing the stratification in deep convection sites. The AMOC then decreases, positive salinity anomalies appear in the Central Arctic, and the variability switches to the opposite phase. These AMOC and sea ice fluctuations have broader climate impacts, with a strong AMOC leading to a mean warming of about 0.4°C north of 20°N, reaching up to 1°C in the Arctic lower troposphere during winter. In all seasons, a northward displacement of the intertropical convergence zone is also simulated. Plain Language Summary: The North Atlantic Ocean is known to have large climate fluctuations emerging from the different components of the climate system and their interactions. These fluctuations play a crucial role in the North American and European climate or the Arctic sea ice. A proper understanding of such internal variations is key to attribute the observed climate changes to anthropogenic activities or to assess the skill of decadal forecast systems. However, investigations of the long‐term basin‐scale variations are restrained by the limited instrumental observations. Therefore, an atmosphere‐ocean general circulation model is used here to explore the low‐frequency variability. This model simulates a large variability in the North Atlantic with a period between 1 century and 1 millennium. We found that this variability owes its existence to the freshwater exchanges between the North Atlantic and Arctic. Such North Atlantic variability has important impacts, as typical positive anomalies of the Atlantic oceanic northward heat transport reduce the sea ice, warm up the whole Northern Hemisphere by 0.4°C, and shift northward the intertropical convergence zone. Key Points: A centennial to multicentennial variability emerges from the North Atlantic and Arctic Ocean in the IPSL‐CM6A‐LR modelAMOC oscillations are driven by freshwater build‐up and release in the ArcticThe interplay between sea ice and oceanic freshwater export modulates the salinity anomalies in the Arctic [ABSTRACT FROM AUTHOR]

A novel decomposition of the ocean heat energy that contributes to sea ice melt and growth (ocean‐ice and frazil heat) into components that are driven by surface heat flux and ocean circulation changes is used to isolate the evolving roles of the atmosphere and ocean in the Arctic sea ice loss from CO2 increases. A sea ice volume budget analysis is used to separate the impacts of the anomalous frazil/ocean‐ice heat from those of atmosphere‐ice heat on the evolving Arctic sea ice volume. The role of atmosphere‐ocean coupling in augmenting or curtailing the atmosphere‐ and ocean‐driven sea ice losses is further isolated by comparing the ice volume budget and the anomalous frazil/ocean‐ice heat components in partially and fully coupled experiments. Atmosphere‐ice heat fluxes drive most of Arctic sea ice loss in the first decade following CO2 increase by increasing the sea ice top face melt in summer, while ocean circulation changes drive the loss over the longer term through the anomalous increase of heat transport into the Arctic, which drive decreases in frazil ice growth and sea ice extent in winter. Atmosphere‐ocean coupling in the subpolar Atlantic further supports a negative feedback that attenuates the ocean‐driven sea ice losses over time; by accelerating the weakening of the Atlantic meridional overturning circulation, it causes a large cooling of the subpolar Atlantic and attenuation of the anomalous heat transport into the Arctic in winter, allowing for a seasonal Arctic sea ice in the fully coupled experiment, while the Arctic completely becomes ice free in the partially coupled experiment. Plain Language Summary: The interactions between the atmosphere, ocean, and sea ice are disentangled in order to isolate their individual roles in driving Arctic sea ice losses and how these roles change over time. We introduce a new decomposition of the ocean heat that contributes to melting and growing sea ice into parts that are caused by changes in the surface heat input into the ocean and ocean circulation. We analyze the contributions of the ocean heat and atmosphere heat to the overall Arctic sea ice volume changes, both in an experiment where a two‐way interaction between the atmosphere and ocean is allowed and in another where this two‐way interaction is not allowed. Our results suggest that the atmosphere causes most of Arctic sea ice loss (within a decade) by increasing sea ice top face melt, while the ocean causes the long‐term sea ice changes in the Arctic by decreasing winter bottom face growth. The interactions between the atmosphere and ocean further enhance the ocean circulation weakening and surface cooling in the subpolar Atlantic and thereby stop the winter heat transport increase into the Arctic and eventually curtail the ocean‐driven sea ice loss and prevent the total loss of Arctic sea ice. Key Points: A novel decomposition isolates how ocean surface heating, circulation changes, and air‐sea interactions drive sea ice loss from CO2 increasesAtmosphere changes determine Arctic sea ice loss within the first decade; ocean circulation changes drive longer term ice lossesAtmosphere‐ocean coupling feedbacks on ocean circulation change in the subpolar Atlantic stabilizes Arctic sea ice loss in the long term [ABSTRACT FROM AUTHOR]

In climate models, the subgrid‐scale orography (SSO) parameterization imposes a blocked flow drag at low levels that is opposed to the local flow. In IPSL‐CM6A‐LR, an SSO lift force is also applied perpendicular to the local flow to account for the effect of locally blocked air in narrow valleys. Using IPSL‐CM6A‐LR sensitivity experiments, it is found that the tuning of both effects strongly impacts the atmospheric circulation. Increasing the blocking and reducing the lift lead to an equatorward shift of the Northern Hemisphere subtropical jet and a reduction of the midlatitude eddy‐driven jet speed. It also improves the simulated synoptic variability, with a reduced storm‐track intensity and increased blocking frequency over Greenland and Scandinavia. Additionally, it cools the polar lower troposphere in boreal winter. Transformed Eulerian Mean diagnostics also show that the low‐level eddy‐driven subsidence over the polar region is reduced consistent with the simulated cooling. The changes are amplified in coupled experiments when compared to atmosphere‐only experiments, as the low‐troposphere polar cooling is further amplified by the temperature and albedo feedbacks resulting from the Arctic sea ice growth. In IPSL‐CM6A‐LR, this corrects the warm winter bias and the lack of sea ice that were present over the Arctic before adjusting the SSO parameters. Our results, therefore, suggest that the adjustment of SSO parameterization alleviates the Arctic sea ice bias in this case. However, the atmospheric changes induced by the parametrized SSO also impact the ocean, with an equatorward shift of the Northern Hemisphere oceanic gyres and a weaker Atlantic meridional overturning circulation. Plain Language Summary: Some of the processes responsible for the impacts of orography on the mean flow, such as low‐level flow blocking, or mountain waves, are unresolved in climate models at standard horizontal resolution. Such processes are accounted for using subgrid‐scale orography parameterization in climate models. Adjusting such parameterization is well known to improve the simulation of the mean climate in midlatitudes and to increase the skill of operational forecasts. In this study, the impact on the Arctic climate is studied in a climate model. It is found that adjusting the subgrid‐scale orography parameterization modulates both the atmospheric variability and mean state, with a large impact on the atmospheric momentum, heat, and moisture transport from the midlatitude to the Arctic. In particular, increasing the low‐level flow blocking leads to decreased atmospheric transport to the Arctic. Such impacts are found in both atmosphere‐only and coupled ocean‐atmosphere sensitivity simulations designed to investigate the influence of the parametrized orography. The coupled climate simulations further illustrate the impact of the subgrid‐scale orography adjustment for the sea ice and oceanic circulation. Increasing low‐level flow blocking is found to increase substantially the winter sea ice growth, while it reduces the Atlantic meridional overturning circulation. Key Points: The adjustment of the parametrized orography in climate models can alleviate the near‐surface winter biases over the ArcticIncreasing low‐level drag reduces Northern Hemisphere stationary wave amplitudes and shifts the subtropical jet equatorwardIncreasing low‐level drag increases the Arctic sea ice coverage and reduces the Atlantic meridional overturning circulation [ABSTRACT FROM AUTHOR]

Sea ice formation processes occur on subgrid scales, and the detailed physics describing the processes are therefore not generally represented in climate models. One likely consequence of this is the premature closing of areas of open water in model simulations, which may result in a misrepresentation of heat and gas exchange between the ocean and atmosphere. This work demonstrates the implementation of a more realistic model of sea ice formation, introducing grease ice as a wind and oceanic stress‐dependent intermediary state between water and new sea ice. We use the fully coupled land‐atmosphere‐ocean‐sea ice model, HadGEM3‐GC3.1 and perform a three‐member ensemble with the new grease ice scheme from 1964 to 2013. Comparing our sea ice results with the existing ensemble without grease ice formation shows an increase in sea ice thickness and volume in the Arctic. In the Antarctic, including grease ice processes results in large local changes to both simulated sea ice concentration and thickness, but no change to the total area or volume. Plain Language Summary: The way that new sea ice forms in most climate models may result in new sea ice forming more quickly than it does in reality, prematurely closing areas of open water that are important to heat and gas exchange between the ocean and atmosphere, and impacting the albedo, and therefore the radiation budget, of the planet. In this work, we implement a more realistic representation of how new sea ice forms in a fully coupled climate model, and demonstrate the effect using an ensemble of historical climate simulations. Key Points: A more detailed representation of sea ice formation was implemented in a coupled climate model ensembleIncluding grease ice processes resulted in increased Arctic sea ice thickness and volumeIncluding grease ice processes resulted in large local changes to Antarctic winter sea ice concentration and thickness [ABSTRACT FROM AUTHOR]