Search

Bu çalışmada amacımız, Güney Okyanusu üzerindeki rüzgar dinamiklerini ve bunun okyanus devinim sirkülasyonu üzerindeki etkisini incelemektir. Bu amaçla, atmosfer ve bileşik okyanus-deniz buzu yüksek çözünürlüklü bölgesel modelleri ayrı ayrı koşturulmuştur. 2007 ve 2013 yılları arasında eşzamanlı olarak üç benzetim gerçekleştirilmiştir. İlk benzetim, gözlemlenen deniz yüzeyi sıcaklığı ve deniz buzu konsantrasyonu tarafından zorlanan sadece atmosfer bölgesel modelidir. Model, ortalama deniz seviyesi basıncı, 2 metre hava sıcaklığı, yukarı atmosfer jetleri ve Stratosferik Polar Vortex gibi önemli atmosferik özelliklerin mevsimselliğini başarıyla yakalamıştır. Model, Antarktika'daki gözlem istasyonlarıyla uyumluluk göstermektedir. İkinci benzetim, reanaliz atmosferik veri seti ile zorlanan kontrol okyanus-deniz buzu bileşik bölgesel modeldir. Okyanus modeli, deniz yüzeyi sıcaklık gradyanını doğru şekilde yakalamayı başarmıştır. Drake Geçidi'ndeki taşınım değerleri gözlemler dahilinde yaklaşık 152 Sv'dir. Son olarak, Güney Okyanusu üzerindeki bölgesel rüzgar gerilmesinin 1,5 kat arttığı bir duyarlılık benzetimi de yapılmış ve daha güçlü Drake Geçidi taşınımı ve Deacon Hücresi sirkülasyonu gözlemlenmiştir. Bu çalışma ileride gerçekleştirilebilecek Güney Okyanusu tamamen bütünleşik atmosfer-okyanus modeli geliştirilmesi için kapasite ve kabiliyetlerin ortaya konmasını sağlamıştır., In this study, our aim is to investigate Southern Ocean wind dynamics and its impact on the ocean overturning circulation. To this end, we performed atmosphere and ocean-sea ice coupled regional high-resolution models separately. We conduct three concurrent simulations spanning between 2007 and 2013. The first simulation is atmosphere only regional model forced by observed sea surface temperature and sea ice concentration. The model successfully captures important atmospheric properties such as mean and seasonality of the sea level pressure, 2 meter air temperature, upper level jet, Stratospheric Polar vortex. The model compares well against the observation stations throughout the Antarctica. The second simulation is the control ocean-sea ice coupled regional model forced with reanalysis atmospheric dataset. In the ocean model, we capture the sea surface temperature gradient. The transport at the Drake Passage is around 152 Sv which is within the observation values. Finally, we conduct a sensitivity simulation where the zonal wind stress over the Southern Ocean is increased 1.5 times. This leads to stronger Drake Passage transport and Deacon Cell overturning circulation in the model. This study has provided to demonstrate the capacity and capabilities to develop a Southern Ocean integrated fully coupled atmosphere-ocean model that can be carried out in the future.

This work presents the last version of the operational analysis and forecasting system of the physical variables and evolution plans in the Black Sea, as developed in the framework of the Copernicus Marine Service for the Black Sea Monitoring and Forecasting Centre. The modelling system consists of NEMO v4.0 hydrodynamical model at 1/40º resolution in horizontal and 121 vertical levels, online coupled to OceanVar for the assimilation of available insitu and satellite observations. Recently, the system has been upgraded to handle the operational historical and forecasting discharge data for the Danube River as provided by the NIHWM (Romania). Major details on the model setup and product are also available at https://resources.marine.copernicus.eu/product-detail/BLKSEA_ANALYSISFORECAST_PHY_007_001/INFORMATION. This contribution will focus on the description of the Black Sea Physics production unit operational capacity and on the accuracy of the delivered products by computing relevant metrics for analysis fields and forecasting skills.

The Mediterranean Basin, including the Mediterranean Sea and the surrounding countries, is referred to as a hotspot in terms of climate change, primarily because of a basin-wide drying trend projected for its future. The Mediterranean Sea plays an important role in the climate of the basin through air-sea interactions, and it is, therefore, important to understand how it is coupled with global as well as regional atmosphere. Coarse resolution fully coupled Earth System Models (ESM) show inaccurate results in terms of sea surface temperature (SST) and precipitation over the Mediterranean Sea and Europe. Better representation of the Mediterranean Sea SST (MedSST) by ESMs is a critical issue for the Euro-Mediterranean climate.In this study, we conduct three simulations using the fully-coupled Community Earth System Model (CESM): i) a historical control simulation integrated for the 1850-2014 period subject to anthropogenic forcings; ii) a Mediterranean Pacemaker-I (MedP-I) experiment where MedSST is nudged to the monthly Extended Reconstructed SST (ERSST) starting from 1880; and iii) a Mediterranean Pacemaker-II (MedP-II) experiment where the MedSST is nudged to the Optimum Interpolation SST (OISST) starting from 1980. In both pacemaker experiments, in comparison with the control simulation, nudging of the MedSST affects the poleward energy flux transported by the atmospheric latent and dry heat, and changes the total meridional energy flux by more than ±0.1 PW over lower latitudes. Similarly, net radiation flux at the surface is changed by about ±2 W/m2 over the Mediterranean Basin. The fidelity of the nudging method was investigated by comparing solutions from MedP-I and MedP-II with respective fields from the control simulation and those from observations, i.e., World Ocean Atlas, Hadley Centre Sea Ice and SST, Climate Prediction Center, and European Observations for the 1981 - 2010 period. The control simulation shows higher surface temperatures than observations and overestimates the total precipitation over Euro-Mediterranean and Turkey. In contrast, both MedP-I and MedP-II show improvements in reproducing total precipitation over the Euro-Mediterranean region, Turkey, and at the entrance of the Gibraltar Strait. While MedP-I has improvements over the northeast Europe and the southern Mediterranean Basin regarding the surface temperatures, MedP-II has some improvements over Turkey and at the coastal areas of the Mediterranean Sea. MedP-II has more improvements for the SST and sea surface salinity (SSS) values over the Mediterranean Sea and the Black Sea compared to MedP-I. Additionally, MedP-II has a better representation of the North Atlantic SSS bias compared to the control simulation, while both MedP-I and MedP-II have some SST improvements for different areas over the North Atlantic. Core climate indices defined by the European Climate Assessment and Dataset project are calculated using simulated daily parameters and results are compared with the Global Land Data Assimilation System dataset. Accordingly, MedP-II is found to have improvements over more areas, especially for the indices calculated by using daily precipitation. Overall, we conclude that Mediterranean Sea Pacemaker simulations improve our understanding of how the Mediterranean Sea impacts the surface temperature and precipitation over the Euro-Mediterranean.

Ocean reanalyses reconstruct the ocean state with a long integration of an ocean model constrained by atmospheric surface forcing and observations via data assimilation. Since their results are more accurate in comparison to those derived from a model alone, they are a powerful tool to provide ocean monitoring indicators (OMIs) as well as to better understand the physical properties and dynamics of the Black Sea. In the scope of the Copernicus Marine Service (CMEMS), the Black Sea physical reanalysis (BS-REA) system has been used to support the implementation of new indicators in the Black Sea. The system is built upon the hydrodynamic model NEMO v3.6 at 1/27° x 1/36°, and 31 unevenly distributed vertical levels, coupled to OceanVar for the assimilation of the best available observations (both in situ and satellite ones). In this contribution, we present the current operational indicators for the monitoring of temperature and salinity anomalies. BS-REA system provides temperature trends that indicate a warming of the basin: 0.0829±0.01069 oC year-1, 0.0380±0.0005 oC year-1, 0.0041±0.0001 oC year-1, respectively, in 0-25 m, 25-150 m and 150-300 m. Since 2007, the warming signal has been very clear in such a way that the Black Sea cold intermediate layer (CIL) almost disappeared in recent years. However, this continuous warming is interrupted in 2012 and less explicitly in 2017, years in which a replenishment of the CIL is verified. Similar analyses for salinity reveal that salinity trends reduce in depth and are larger from 2005, especially in surface layers. The system is very suitable for understanding the physical state of the Black Sea in recent years and allows to obtain more accurate OMIs for the sea, which are important to understand its response to climate change. Following recent CMEMS Ocean State Report contributions, new OMIs based on the ocean heat content anomaly and freshwater content anomaly will be produced this year, whereas indexes on the basis of the interannual variations of the Black Sea Rim Current intensity, the Black Sea overturning circulation and the coastal upwelling along the Turkish coast will be provided as OMIs in the near future. BS-REA is under continuous update in order that new versions will bring improvements in both the numerical model and data assimilation components.

The Black Sea Monitoring and Forecasting Center (BS-MFC) is the European reference service for the provision of ocean analyses, forecasts, and reanalyses in the Black Sea basin. It is part of the Copernicus Marine Environment and Monitoring Service (CMEMS) and ensures a high level of efficiency in terms of operations, science, and technology for predictions and the monitoring of physical and biogeochemical processes in the Black Sea. The operational BS-MFC framework is based on state-of-the-art numerical models for hydrodynamics, biogeochemistry, and waves, analysis, forecast, and reanalysis are provided on a spatial grid with about 3 km of horizontal resolution that covers the whole Black Sea basin (the Azov Sea is not included). The scientific assessment of BS-MFC products is performed by implementing a product quality dashboard that provides pre-qualification and operational model skills according to GODAE/OceanPredict standards. Novel interfaces based on high-resolution models are part of the scientific development plan to ensure a strong connection with the nearest seas from a modelling point of view, in particular with the Mediterranean Sea. To improve forecasting skills, dedicated online coupled systems are being developed, which involve physics, biogeochemistry, and waves together with the atmosphere and, in the future, with ensemble forecasting methodologies and river-ocean interfaces.

A new computationally efficient version of the Norwegian Earth System Model (NorESM) is presented. This new version (here termed NorESM1-F) runs about 2.5 times faster (e.g., 90 model years per day on current hardware) than the version that contributed to the fifth phase of the Coupled Model Intercomparison project (CMIP5), i.e., NorESM1-M, and is therefore particularly suitable for multimillennial paleoclimate and carbon cycle simulations or large ensemble simulations. The speed-up is primarily a result of using a prescribed atmosphere aerosol chemistry and a tripolar ocean–sea ice horizontal grid configuration that allows an increase of the ocean–sea ice component time steps. Ocean biogeochemistry can be activated for fully coupled and semi-coupled carbon cycle applications. This paper describes the model and evaluates its performance using observations and NorESM1-M as benchmarks. The evaluation emphasizes model stability, important large-scale features in the ocean and sea ice components, internal variability in the coupled system, and climate sensitivity. Simulation results from NorESM1-F in general agree well with observational estimates and show evident improvements over NorESM1-M, for example, in the strength of the meridional overturning circulation and sea ice simulation, both important metrics in simulating past and future climates. Whereas NorESM1-M showed a slight global cool bias in the upper oceans, NorESM1-F exhibits a global warm bias. In general, however, NorESM1-F has more similarities than dissimilarities compared to NorESM1-M, and some biases and deficiencies known in NorESM1-M remain.

The Arctic Ocean receives a net freshwater input from land and from the atmosphere. This flux of freshwater, along with net surface heat loss, acts to transform the water mass properties of inflowing Atlantic and Pacific waters. In this study, model simulations are used to quantify the Arctic water mass transformation in salinity and temperature space, and its explained variance due to variability in the largest freshwater source to the Arctic: river runoff. This explained variance is quantified using a novel tool, the seasonal climate response function, which describes the magnitude and time scale of adjustment to a runoff perturbation at monthly resolution. Using this method, the transient response of Arctic water mass transformation is reconstructed over time scales ranging from several months to a decade. Model simulations with variable runoff indicate a significant explained model variance of several terms contributing to salinity transformation, including diffusion, the formation and melt of sea ice, and a possibly model‐dependent surface salinity‐restoring term. Most notably, an increase in river runoff strengthens the diffusion of salt and heat, which ultimately leads to an increase in the advective salt and heat import into the Arctic. These results provide evidence for the potential predictability of the Arctic system based on variability in river runoff. publishedVersion

International audience; Copernicus Marine Service (CMEMS) includes the Black Sea Monitoring and Forecasting Center (BS-MFC) for the provisioning of high-quality forecast products and past reconstruction of the ocean state in the Black Sea. The BS-MFC is implemented by an international consortium since 2016, led by IO-BAS (Bulgaria) and including CMCC (Italy), ULiege (Belgium), HZ Hereon (Germany), USOF (Bulgaria), NIHWM (Romania), together with expertise from University of Bologna (Italy), ITU (Turkey) and DEU (Turkey). For each Production Unit-Physics, Biogeochemistry and Waves-an operational service and scientifi c evolution plans have been implemented. Every day, BS-MFC delivers analysis and 10-days forecast products for essential variables, including biogeochemistry and waves, at the spatial resolution of about 3 km. NRT systems are forced by ECMWF IFS atmospheric forcing, at 12.5 km horizontal resolution and 3-6 hours frequency in time. Twice per year, BS-MFC delivers MY products (reanalysis) starting from 1992 (1979 for BS-WAV). Systems are forced by ECMWF ERA5 atmospheric forcing, at 30 km horizontal resolution and 1-hour frequency in time. Scientifi c plans for the next generation of the BS-MFC systems put in top priority coupling strategy, ensemble forecasting, data assimilation capability and river runoff representation to support downstream services and applications, climate monitoring and coastal community.

International audience; This study will present the evolution of the Black Sea model physics in the framework of the Copernicus Marine Environment Monitoring Service, from EAS3 (current operational system) to EAS4 (operational from May 2021). EAS4 has been greatly improved thanks to new core model (based on NEMO v4.0), improved bathymetry and representation of rivers, revision of the spatial domain at higher resolution (2.5 km and 121 z-star-levels) to include the Marmara Sea box for the provision of the lateral open boundary conditions. For the optimal interface between Mediterranean and Black Sea though boundary conditions, a new high-resolution unstructured grid-based model has been developed for the Marmara Sea. Evolution in terms of modelling framework and prediction capacity will be detailed, describing the main results and skill score metrics.

The Black Sea physical analysis and Forecasting System (BSFS) is part of the Black Sea Monitoring and Forecasting Centre (BS-MFC) for the Copernicus Marine Service (CMEMS). It provides analysis every day analysis and 10 days forecast fields for the blue ocean variables (including temperature, salinity, sea surface height, mixed layer depth and currents) in the Black Sea region since. In this work, we present the new version of the operational system that will be part of the next CMEMS release. The hydrodynamical core model is based on NEMO v4.0, solved on 1/40º horizontal resolution spatial grid (including the overall Black Sea, the Bosporus Strait and part of the Marmara Sea) and 121 vertical levels with z-star. The core model uses ECMWF analysis and forecast atmospheric forcing and GPCP monthly climatological precipitation for computing heat, water and momentum fluxes. A total number of 72 rivers is accounted, as monthly climatology provided by SESAME project. The model implements a new representation of the Danube River with interannual river discharge datasets provided by the National Institute of Hydrology and Water Management. One of the main innovations of this system is the opening of the Bosporus Strait by using a box-approach in a portion of the Marmara Sea: it is achieved thanks to high resolution temperature, salinity, sea surface height, zonal and meridional velocity solutions provided by a novel implementation of the Marmara Sea model including straits based on Shyfem: it represents the optimal interface between the Mediterranean and the Black Sea. The hydrodynamical model is online coupled to an upgraded version of the OceanVar, the CMCC data assimilation scheme, able to assimilate SLA L3 satellite data, T/S in-situ profiles and SST from CMEMS TACs. The contribution focuses on model setup description, processing system and validation. To evaluate BSFS pre-operational run and monitor the operational production, we provide metrics as proposed within GODAE/Oceanpredict and MERSEA/MyOcean (which includes CLASS 1, 2 and 4 metrics).

Setting new model configurations based on NEMO requires the definition of initial/boundary condition and the validation of numerical solutions. In the framework of IMMERSE H2020 project, CMCC is developing new tools and technological capacities for handling in easy and reliable way external products, such CMEMS or coastal ocean data, for research-to-operations applications. Generic Interfaces for NEMO (InterNEMO) allow for 3 main scopes: 1) to access and discover the CMEMS catalogue, including both model and observational data; 2) to manipulate accessed datasets, including coastal ocean data, to extract relevant physical information to use for setting initial/boundary conditions for a new NEMO-based configurations; 3) to prepare NEMO set of upstream files and to validate NEMO solution by using CMEMS observational datasets. InterNEMO implements also technologies to connect a NEMO user to Wekeo DIAS (https://www.wekeo.eu/) for the interoperable accessing and processing of CMEMS data. In this contribution, we present the InterNEMO architecture developed in Python via Jupyter Notebooks, to support the user/researcher to easily discover, design and configure modeling components required by the new NEMO-based configuration. InterNEMO is tested for the Black Sea hydrodynamical model configuration, developed by CMCC in the framework of the Black Sea Monitoring and Forecasting Centre (BS-MFC) for CMEMS a) to show how to access CMEMS observations through Wekeo DIAS and use them to validate numerical solutions and b) to define open boundary conditions from an unstructured grid model configuration based on Shyfem, developed for the Marmara Sea.

Marmara Sea including Bosphorus and Dardanelles Straits (i.e. Turkish Strait Systems, TSS) is the connection between the Black Sea and the Mediterranean. The exchange flow that occurs in the Straits is crucial to set the deep water properties in the Black Sea and the surface water conditions in the Northern Aegean Sea. We have developed a new high-resolution unstructured grid model (U-TSS) for the Marmara Sea including the Bosporus and Dardanelles Straits using the System of HydrodYnamic Finite Element Modules (SHYFEM). Using an unstructured grid in the horizontal better resolves geometry of the Turkish Straits. The new model has a resolution between 500 meter in the deep to 50 meter in the shallow areas, and 93 geopotential coordinate levels in the vertical. We conducted a 4 year hindcast simulation between 2016 and 2019 using lateral boundary conditions from CMEMS (https://marine.copernicus.eu/) analysis, in particular Black Sea Forecasting System (BS-FS) for the northern boundary and Mediterranean Sea Forecasting System (MS-FS) for the southern boundary. Atmospheric boundary conditions fare from the ECMWF dataset.Mean averaged surface circulation shows that there is a cyclonic gyre in the middle of the basin due to Bosphorus jet flowing to the south and turning to west after reaching the southern Marmara coast. The U-TSS model has been validated against the seasonal in situ observations obtained from four different cruises between 2017 and 2018. The maximum bias occurs at around halocline depth between 20 to 30 meters. We also found that root mean square error field is higher in the mixed layer interface. We conclude that capturing shallow mixed layer depth is very in the Marmara Sea due to the interplay of air-sea fluxes and mixing parametrizations uncertainties. Maximum salinity bias and rms in the new U-TSS model are around 3 psu which is a significant improvement with respect to previous studies. This new model will be used as an operational forecasting system and will provide lateral boundary conditions for the BS-FS and MS-FS models in the future.

In this study, a high-resolution eddy resolving regional ocean + sea ice coupled model (MITgcm) is used to study the effects of increasing westerlies along the Southern Ocean. Previous studies only focused on increasing wind stress, thus not taking into account of atmosphere-to-ocean heat and freshwater fluxes. Here, we conduct two concurrent simulations; i) 1.5 times increased wind stress (i.e. increased only mechanical forcing) ii) 1.2247 times increased wind speed (i.e. both mechanical and thermal flux forcing). Model domain covers whole Southern Hemisphere with lateral open boundary conditions from ECCOv2 ocean reanalysis and surface boundary conditions from ECMWF ERA-5 atmospheric reanalysis. In both sensitivity scenarios, due to the increase in the wind stress, the Ekman transport towards Equator towards north is increased. This caused increased upwelling of warmer North Atlantic Deep Water (NADW) near the Antarctic ice sheet. Both scenarios show reduced sea ice formation with up to 2 million km2 in the austral summer and up to 4 million km2 during the austral winter. Sea ice extent is reduced more in the mechanical forcing simulation than the mechanical+thermal forcing one. This is a clear result that increased wind anomalies should be studied with increased wind rather than increased stress. The reduction in the sea ice coverage that is attributed to the warmer water mass can also be observed through the Sea Surface Temperature (SST) values. The first case shows up to 1 – 1.5 °C very close to the Antarctica, whereas the second case shows a much limited SST change around 0.5 °C.Both sensitivity scenarios show an increase of the transport along Drake Passage. However, the mechanical+thermal case shows larger increase in the Drake transport compared to the mechanical case. This indicates that a change in the Antarctic Circumpolar Circulation also modifies the meridional density gradient along with the upwelling characteristics. Finally, overturning transport in the density space shows that Subtropical Cell and ACC upper Cell strengthen in the mechanical+thermal case, while there are no significant changes in the thermal case. In both simulations, Subpolar Cell increases and Lower Cell decreases. We conclude that studying increased westerlies with two different approaches show significant changes in the surface and deep circulation. Previous studies which taken into only mechanical forcing part are missing thermal component of the wind effects.

The second version of the coupled Norwegian Earth System Model (NorESM2) is presented and evaluated. NorESM2 is based on the second version of the Community Earth System Model (CESM2) and shares with CESM2 the computer code infrastructure and many Earth system model components. However, NorESM2 employs entirely different ocean and ocean biogeochemistry models. The atmosphere component of NorESM2 (CAM-Nor) includes a different module for aerosol physics and chemistry, including interactions with cloud and radiation; additionally, CAM-Nor includes improvements in the formulation of local dry and moist energy conservation, in local and global angular momentum conservation, and in the computations for deep convection and air–sea fluxes. The surface components of NorESM2 have minor changes in the albedo calculations and to land and sea-ice models. We present results from simulations with NorESM2 that were carried out for the sixth phase of the Coupled Model Intercomparison Project (CMIP6). Two versions of the model are used: one with lower (∼ 2∘) atmosphere–land resolution and one with medium (∼ 1∘) atmosphere–land resolution. The stability of the pre-industrial climate and the sensitivity of the model to abrupt and gradual quadrupling of CO2 are assessed, along with the ability of the model to simulate the historical climate under the CMIP6 forcings. Compared to observations and reanalyses, NorESM2 represents an improvement over previous versions of NorESM in most aspects. NorESM2 appears less sensitive to greenhouse gas forcing than its predecessors, with an estimated equilibrium climate sensitivity of 2.5 K in both resolutions on a 150-year time frame; however, this estimate increases with the time window and the climate sensitivity at equilibration is much higher. We also consider the model response to future scenarios as defined by selected Shared Socioeconomic Pathways (SSPs) from the Scenario Model Intercomparison Project defined under CMIP6. Under the four scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5), the warming in the period 2090–2099 compared to 1850–1879 reaches 1.3, 2.2, 3.0, and 3.9 K in NorESM2-LM, and 1.3, 2.1, 3.1, and 3.9 K in NorESM-MM, robustly similar in both resolutions. NorESM2-LM shows a rather satisfactory evolution of recent sea-ice area. In NorESM2-LM, an ice-free Arctic Ocean is only avoided in the SSP1-2.6 scenario.

The second version of the fully coupled Norwegian Earth System Model (NorESM2) is presented and evaluated. NorESM2 is based on the second version of the Community Earth System Model (CESM2), but has entirely different ocean and ocean biogeochemistry models; a new module for aerosols in the atmosphere model along with aerosol-radiation-cloud interactions and changes related to the moist energy formulation, deep convection scheme and angular momentum conservation; modified albedo and air-sea turbulent flux calculations; and minor changes to land and sea ice models. We show results from low (∼2°) and medium (∼1°) atmosphere-land resolution versions of NorESM2 that have both been used to carry out simulations for the sixth phase of the Coupled Model Intercomparison Project (CMIP6). The stability of the pre-industrial climate and the sensitivity of the model to abrupt and gradual quadrupling of CO2 is assessed, along with the ability of the model to simulate the historical climate under the CMIP6 forcings. As compared to observations and reanalyses, NorESM2 represents an improvement over previous versions of NorESM in most aspects. NorESM2 is less sensitive to greenhouse gas forcing than its predecessors, with an equilibrium climate sensitivity of 2.5 K in both resolutions on a 150 year frame. We also consider the model response to future scenarios as defined by selected shared socioeconomic pathways (SSP) from the Scenario Model Intercomparison Project defined under CMIP6. Under the four scenarios SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5, the warming in the period 2090–2099 compared to 1850–1879 reaches 1.3, 2.2, 3.0, and 3.9 K in NorESM2-LM, and 1.3, 2.1, 3.1, and 3.9 K in NorESM–MM, robustly similar in both resolutions. NorESM2-LM shows a rather satisfactorily evolution of recent sea ice area. In NorESM2-LM an ice free Arctic Ocean is only avoided in the SSP1-2.6 scenario.

Highlights: • Assessment of the Indian Ocean simulation from global forced sea- ice models. • SST biases are 2 times smaller in forced simulations than the coupled simulations. • Coupled model shows large inter-model spread over the eastern equatorial Indian Ocean. • Refinement in model horizontal resolution does not significantly improve simulations. • Uncover a secondary pathway of northward cross-equatorial transport along 75 °E. • Models are unable to capture the observed thick barrier layer in the north Bay of Bengal. Abstract: We present an analysis of annual and seasonal mean characteristics of the Indian Ocean circulation and water masses from 16 global ocean–sea-ice model simulations that follow the Coordinated Ocean-ice Reference Experiments (CORE) interannual protocol (CORE-II). All simulations show a similar large-scale tropical current system, but with differences in the Equatorial Undercurrent. Most CORE-II models simulate the structure of the Cross Equatorial Cell (CEC) in the Indian Ocean. We uncover a previously unidentified secondary pathway of northward cross-equatorial transport along 75 °E, thus complementing the pathway near the Somali Coast. This secondary pathway is most prominent in the models which represent topography realistically, thus suggesting a need for realistic bathymetry in climate models. When probing the water mass structure in the upper ocean, we find that the salinity profiles are closer to observations in geopotential (level) models than in isopycnal models. More generally, we find that biases are model dependent, thus suggesting a grouping into model lineage, formulation of the surface boundary, vertical coordinate and surface salinity restoring. Refinement in model horizontal resolution (one degree versus degree) does not significantly improve simulations, though there are some marginal improvements in the salinity and barrier layer results. The results in turn suggest that a focus on improving physical parameterizations (e.g. boundary layer processes) may offer more near-term advances in Indian Ocean simulations than refined grid resolution.

The Turkish Strait System, which is the only connection between the Black Sea and the Mediterranean Sea, is a challenging region for ocean circulation models due to topographic constraints and water mass structure. We present a newly developed high resolution unstructured finite element grid model to simulate the Turkish Strait System using realistic atmospheric forcing and lateral open boundary conditions. We find that the jet flowing from the Bosphorus Strait into the Marmara creates an anticyclonic circulation. The eddy kinetic energy field is high around the jets exiting from the Bosphorus Strait, Dardanelles Strait, and also the leeward side of the islands in the Marmara Sea. The model successfully captures the two-layer structure of the Sea of Marmara. The volume transport at the Bosphorus is around 120 km3/year which is consistent with the recent observations. The largest bias in the model is at the interface depth due to the shallower mixed layer.

This work focuses on enabling the use of existing in situ analysis and visualization methods within the model coupling framework, which is specially designed software layer to combine physical components of Earth system (i.e., atmosphere, ocean) through a standardized calling interface to simplify common tasks such as interpolation among different numerical grids and synchronization of the model components. The new approach enables simultaneous analysis of the vast amount of data produced by a multi-component Earth system model in an integrated way and facilitates efficient, interoperable, and optimized data analysis and visualization workflows for Earth system science applications. To demonstrate the flexibility and the capability of the new approach, a fully coupled regional atmosphere-ocean modeling application that aims to investigate air-sea interaction in a very high spatial and temporal resolution has been implemented. In the use case application, both the simulation and in situ visualization application run in parallel on a large-scale HPC platform with both software and hardware rendering support through the use of accelerators.

Plain Language Summary: The North Atlantic Current is an extension of the Gulf Stream, which brings warm Atlantic Water northward as the current flows through the Nordic Seas. Eventually, it enters the cold deep Arctic Ocean basins through the Barents Sea and Fram Strait. Nine different numerical ocean ice models have been analyzed and compared in order to investigate (1) their ability to simulate this northward flow of Atlantic Water, (2) its dependence on wind forcing, and (3) its impact on Arctic sea ice. Consistently, in all models, stronger winds in the Greenland Sea result in a stronger northward flow of warm Atlantic Water. The response on ocean circulation occurs from the North Atlantic, through the Nordic Seas and the Barents Sea, to the deep Canadian Basin. The flow of warm Atlantic Water within the Arctic Ocean is thus highly dependent on the wind stress in the Nordic Seas. There is particularly clear response in the Barents and Kara Seas where a wind‐driven anomalous warm inflow drives a smaller sea ice extent and thickness, and an increased heat transfer from the ocean to the atmosphere above. Weaker winds in the Greenland Sea produces weaker flow and hence a larger sea ice extent and thickness. Abstract: Multimodel Arctic Ocean “climate response function” experiments are analyzed in order to explore the effects of anomalous wind forcing over the Greenland Sea (GS) on poleward ocean heat transport, Atlantic Water (AW) pathways, and the extent of Arctic sea ice. Particular emphasis is placed on the sensitivity of the AW circulation to anomalously strong or weak GS winds in relation to natural variability, the latter manifested as part of the North Atlantic Oscillation. We find that anomalously strong (weak) GS wind forcing, comparable in strength to a strong positive (negative) North Atlantic Oscillation index, results in an intensification (weakening) of the poleward AW flow, extending from south of the North Atlantic Subpolar Gyre, through the Nordic Seas, and all the way into the Canadian Basin. Reconstructions made utilizing the calculated climate response functions explain ∼50% of the simulated AW flow variance; this is the proportion of variability that can be explained by GS wind forcing. In the Barents and Kara Seas, there is a clear relationship between the wind‐driven anomalous AW inflow and the sea ice extent. Most of the anomalous AW heat is lost to the atmosphere, and loss of sea ice in the Barents Sea results in even more heat loss to the atmosphere, and thus effective ocean cooling. Release of passive tracers in a subset of the suite of models reveals differences in circulation patterns and shows that the flow of AW in the Arctic Ocean is highly dependent on the wind stress in the Nordic Seas. publishedVersion

Highlights • We propose a new diagnostic method for spurious mixing. • The new method diagnoses the regions of explicit and implicit mixing. • High shear areas are the main regions which are susceptible to numerical truncation errors. • Smagorinsky viscosity has less numerical mixing than the Leith viscosity closure., Numerical mixing is inevitable for ocean models due to tracer advection schemes. Until now, there is no robust way to identify the regions of spurious mixing in ocean models. We propose a new method to compute the spatial distribution of the spurious diapycnic mixing in an ocean model. This new method is an extension of available potential energy density method proposed by Winters and Barkan (2013). We test the new method in lock-exchange and baroclinic eddies test cases. We can quantify the amount and the location of numerical mixing. We find high-shear areas are the main regions which are susceptible to numerical truncation errors. We also test the new method to quantify the numerical mixing in different horizontal momentum closures. We conclude that Smagorinsky viscosity has less numerical mixing than the Leith viscosity using the same non-dimensional constant.

Global ocean models with an isopycnic vertical coordinate are advantageous in representing overflows, as they do not suffer from topography-induced spurious numerical mixing commonly seen in geopotential coordinate models. In this paper, we present a quantitative diagnosis of the Nordic Seas overflows in four configurations of the Norwegian Earth System Model (NorESM) family that features an isopycnic ocean model. For intercomparison, two coupled ocean-sea ice and two fully coupled (atmosphere-land-ocean-sea ice) experiments are considered. Each pair consists of a (non-eddying) 1° and a (eddy-permitting) 1/4° horizontal resolution ocean model. In all experiments, overflow waters remain dense and descend to the deep basins, entraining ambient water en route. Results from the 1/4° pair show similar behavior in the overflows, whereas the 1° pair show distinct differences, including temperature/salinity properties, volume transport (Q), and large scale features such as the strength of the Atlantic Meridional Overturning Circulation (AMOC). The volume transport of the overflows and degree of entrainment are underestimated in the 1° experiments, whereas in the 1/4° experiments, there is a two-fold downstream increase in Q, which matches observations well. In contrast to the 1/4° experiments, the coarse 1° experiments do not capture the inclined isopycnals of the overflows or the western boundary current off the Flemish Cap. In all experiments, the pathway of the Iceland-Scotland Overflow Water is misrepresented: a major fraction of the overflow proceeds southward into the West European Basin, instead of turning westward into the Irminger Sea. This discrepancy is attributed to excessive production of Labrador Sea Water in the model. The mean state and variability of the Nordic Seas overflows have significant consequences on the response of the AMOC, hence their correct representations are of vital importance in global ocean and climate modelling.

The Arctic Ocean simulated in 14 global ocean-sea ice models in the framework of the Coordinated Ocean-ice Reference Experiments, phase II (CORE-II) is analyzed in this study. The focus is on the Arctic liquid freshwater (FW) sources and freshwater content (FWC). The models agree on the interannual variability of liquid FW transport at the gateways where the ocean volume transport determines the FW transport variability. The variation of liquid FWC is induced by both the surface FW flux (associated with sea ice production) and lateral liquid FW transport, which are in phase when averaged on decadal time scales. The liquid FWC shows an increase starting from the mid-1990s, caused by the reduction of both sea ice formation and liquid FW export, with the former being more significant in most of the models. The mean state of the FW budget is less consistently simulated than the temporal variability. The model ensemble means of liquid FW transport through the Arctic gateways compare well with observations. On average, the models have too high mean FWC, weaker upward trends of FWC in the recent decade than the observation, and low consistency in the temporal variation of FWC spatial distribution, which needs to be further explored for the purpose of model development.

A new computationally efficient version of the Norwegian Earth System Model (NorESM) is presented. This new version (here termed NorESM1-F) runs about 2.5 times faster (e.g. 90 model years per day on current hardware) than the version that contributed to the fifth phase of the Coupled Model Intercomparison project (CMIP5), i.e., NorESM1-M, and is therefore particularly suitable for multi-millennial paleoclimate and carbon cycle simulations or large ensemble simulations. The speedup is primarily a result of using a prescribed atmosphere aerosol chemistry and a tripolar ocean-sea ice horizontal grid configuration that allows an increase of the ocean-sea ice component time steps. Ocean biogeochemistry can be activated for fully coupled and semi-coupled carbon cycle applications. This paper describes the model and evaluates its performance using observations and NorESM1-M as benchmarks. The evaluation emphasises model stability, important large-scale features in the ocean and sea ice components, internal variability in the coupled system, and climate sensitivity. Simulation results from NorESM1-F in general agree well with observational estimates, and show evident improvements over NorESM1-M, for example, in the strength of the meridional overturning circulation and sea ice simulation, both important metrics in simulating past and future climates. Whereas NorESM1-M showed a slight global cool bias in the upper oceans, NorESM1-F exhibits a global warm bias. In general, however, NorESM1-F has more similarities than dissimilarities compared to NorESM1-M, and some biases and deficiencies known in NorESM1-M remain.

We present a new surface-atmospheric dataset for driving ocean–sea-ice models based on Japanese 55-year atmospheric reanalysis (JRA-55), referred to here as JRA55-do. The JRA55-do dataset aims to replace the CORE interannual forcing version 2 (hereafter called the CORE dataset), which is currently used in the framework of the Coordinated Ocean-ice Reference Experiments (COREs) and the Ocean Model Intercomparison Project (OMIP). A major improvement in JRA55-do is the refined horizontal grid spacing (∼ 55 km) and temporal interval (3 hr). The data production method for JRA55-do essentially follows that of the CORE dataset, whereby the surface fields from an atmospheric reanalysis are adjusted relative to reference datasets. To improve the adjustment method, we use high-quality products derived from satellites and from several other atmospheric reanalysis projects, as well as feedback on the CORE dataset from the ocean modelling community. Notably, the surface air temperature and specific humidity are adjusted using multi-reanalysis ensemble means. In JRA55-do, the downwelling radiative fluxes and precipitation, which are affected by an ambiguous cloud parameterisation employed in the atmospheric model used for the reanalysis, are based on the reanalysis products. This approach represents a notable change from the CORE dataset, which imported independent observational products. Consequently, the JRA55-do dataset is more self-contained than the CORE dataset, and thus can be continually updated in near real-time. The JRA55-do dataset extends from 1958 to the present, with updates expected at least annually. This paper details the adjustments to the original JRA-55 fields, the scientific rationale for these adjustments, and the evaluation of JRA55-do. The adjustments successfully corrected the biases in the original JRA-55 fields. The globally averaged features are similar between the JRA55-do and CORE datasets, implying that JRA55-do can suitably replace the CORE dataset for use in driving global ocean–sea-ice models.

We characterise the representation of the Southern Ocean water mass structure and sea ice within a suite of 15 global ocean-ice models run with the Coordinated Ocean-ice Reference Experiment Phase II (CORE-II) protocol. The main focus is the representation of the present (1988-2007) mode and intermediate waters, thus framing an analysis of winter and summer mixed layer depths; temperature, salinity, and potential vorticity structure; and temporal variability of sea ice distributions. We also consider the interannual variability over the same 20 year period. Comparisons are made between models as well as to observation-based analyses where available. The CORE-II models exhibit several biases relative to Southern Ocean observations, including an underestimation of the model mean mixed layer depths of mode and intermediate water masses in March (associated with greater ocean surface heat gain), and an overestimation in September (associated with greater high latitude ocean heat loss and a more northward winter sea-ice extent). In addition, the models have cold and fresh/warm and salty water column biases centred near 50 degrees S. Over the 1933-2007 period, the CORE-II models consistently simulate spatially variable trends in sea-ice concentration, surface freshwater fluxes, mixed layer depths, and 200-700 in ocean heat content. In particular, sea-ice coverage around most of the Antarctic continental shelf is reduced, leading to a cooling and freshening of the near surface waters. The shoaling of the mixed layer is associated with increased surface buoyancy gain, except in the Pacific where sea ice is also influential. The models are in disagreement, despite the common CORE-II atmospheric state, in their spatial pattern of the 20-year trends in the mixed layer depth and sea-ice. (C) 2015 Elsevier Ltd. All rights reserved

In the framework of the second phase of the Coordinated Ocean-ice Reference Experiments (CORE-II), we present an analysis of the representation of the Antarctic Circumpolar Current (ACC) and Southern Ocean meridional overturning circulation (MOC) in a suite of seventeen global ocean-sea ice models. We focus on the mean, variability and trends of both the ACC and MOC over the 1958-2007 period, and discuss their relationship with the surface forcing. We aim to quantify the degree of eddy saturation and eddy compensation in the models participating in CORE-II, and compare our results with available observations, previous fineresolution numerical studies and theoretical constraints. Most models show weak ACC transport sensitivity to changes in forcing during the past five decades, and they can be considered to be in an eddy saturated regime. Larger contrasts arise when considering MOC trends, with a majority of models exhibiting significant strengthening of the MOC during the late 20th and early 21st century. Only a few models show a relatively small sensitivity to forcing changes, responding with an intensified eddy-induced circulation that provides some degree of eddy compensation, while still showing considerable decadal trends. Both ACC and MOC interannual variabilities are largely controlled by the Southern Annular Mode (SAM). Based on these results, models are clustered into two groups. Models with constant or two-dimensional (horizontal) specification of the eddy-induced advection coefficient K show larger ocean interior decadal trends, larger ACC transport decadal trends and no eddy compensation in the MOC. Eddy-permitting models or models with a threedimensional time varying K show smaller changes in isopycnal slopes and associated ACC trends, and partial eddy compensation. As previously argued, a constant in time or space lc is responsible for a poor representation of mesoscale eddy effects and cannot properly simulate the sensitivity of the ACC and MOC to changing surface forcing. Evidence is given for a larger sensitivity of the MOC as compared to the ACC transport, even when approaching eddy saturation. Future process studies designed for disentangling the role of momentum and buoyancy forcing in driving the ACC and MOC are proposed. (C) 2015 Elsevier Ltd. All rights reserved.

We have investigated the causes of the sea surface temperature (SST) bias in the Angola–Benguela Frontal Zone (ABFZ) of the southeastern Atlantic Ocean simulated by the Norwegian Earth System Model (NorESM). Similar to other coupled-models, NorESM has a warm SST bias in the ABFZ of up to 8 °C in the annual mean. Our analysis of NorESM reveals that a cyclonic surface wind bias over the ABFZ drives a locally excessively strong southward (0.05 m/s (relative to observation)) Angola Current displacing the ABFZ southward. A series of uncoupled stand-alone atmosphere and ocean model simulations are performed to investigate the cause of the coupled model bias. The stand-alone atmosphere model driven with observed SST exhibits a similar cyclonic surface circulation bias; while the stand-alone ocean model forced with the reanalysis data produces a warm SST in the ABFZ with a magnitude approximately half of that in the coupled NorESM simulation. An additional uncoupled sensitivity experiment shows that the atmospheric model’s local negative surface wind curl generates anomalously strong Angola Current at the ocean surface. Consequently, this contributes to the warm SST bias in the ABFZ by 2 °C (compared to the reanalysis forced simulation). There is no evidence that local air-sea feedbacks among wind stress curl, SST, and sea level pressure (SLP) affect the ABFZ SST bias. Turbulent surface heat flux differences between coupled and uncoupled experiments explain the remaining 2 °C warm SST bias in NorESM. Ocean circulation, upwelling and turbulent heat flux errors all modulate the intensity and the seasonality of the ABFZ errors. publishedVersion

Mixing efficiency between different water masses is typically assumed to be constant in diapycnal mixing parameterizations used in ocean models. As of now, most coarse resolution ocean circulation models employ a constant mixing efficiency value of 0.2 for the shear driven mixing, internal waves and bottom boundary layer parameterizations. This study investigates the energetics and mixing efficiency of the lock-exchange flow at different Reynolds numbers. The lock-exchange experiment resolves Kelvin–Helmholtz vortices and is an idealized test case for oceanic gravity currents. At first, the required spatial resolution for the direct numerical simulations (DNS) is determined in simulations at a constant Reynolds number of 3500. The evolution of background potential energy and tracer variance are used to assess model results. We found that the model spatial resolution should resolve at least the Kolmogorov scale but not necessarily the Batchelor scale if convergences of background potential energy, tracer variance and dissipation are considered. Simulations at Reynolds number of 125, 500, 1000, 2500, 3500, 5000, 6000, 10,000 show that the mixing efficiency in the lock-exchange flow is smaller than 0.2, and it saturates around 0.12 when Reynolds numbers exceed the value of 2500.

The generation mechanism of mesoscale eddies in the Faroe Bank Channel (FBC) overflow region and their spatiotemporal characteristics are examined using the high-resolution regional Massachusetts Institute of Technology general circulation model (MITgcm). From the modeled overflow, it is found that the volume transport downstream of the FBC sill exhibits strong variability with a distinct period of ~4 days. Energetic, alternating cyclonic and anticyclonic eddies appear at ~40 km downstream of the sill. They grow side by side in the nascent stage, but later the cyclones migrate along the 800-m isobath to the south of Iceland, whereas the anticyclones descend downslope across the isobath and gradually dissipate. Analysis of the eddy characteristics shows that the cyclones are associated with a larger plume thickness and width, larger volume transport, colder and denser water, and a plume core located farther downslope, whereas the opposite is true for the anticyclones. The oscillatory structure developed at the lower boundary of the mean plume and the following generation of alternating cyclones and anticyclones are typical features of baroclinic instability. A linear instability analysis of a two-layer analytical baroclinic model yields a most unstable mode that agrees favorably with the simulations. The calculation of the divergent eddy heat flux shows a substantial rightward (upslope)-directed component downstream of the FBC sill. This region is also associated with a strong baroclinic conversion rate. The above arguments constitute evidence for the generation of unstable plume and mesoscale eddies in the FBC region by baroclinic instability.

We propose a new framework for parameterization of ocean convection processes. The new framework is termed “patchy convection” since our aim is to represent the heterogeneity of mixing processes that take place within the horizontal scope of a grid cell. We focus on applying this new scheme to represent the effect of pre-conditioning for deep convection by subgrid scale eddy variability. The new parameterization separates the grid-cell into two regions of different stratification, applies convective mixing separately to each region, and then recombines the density profile to produce the grid-cell mean density profile. The scheme depends on two parameters: the areal fraction of the vertically-mixed region within the horizontal grid cell, and the density difference between the mean and the unstratified profiles at the surface. We parameterize this density difference in terms of an unresolved eddy kinetic energy. We illustrate the patchy parameterization using a 1D idealized convection case before evaluating the scheme in two different global ocean-ice simulations with prescribed atmospheric forcing; (i) diagnosed eddy velocity field applied only in the Labrador Sea (ii) diagnosed global eddy velocity field. The global simulation results indicate that the patchy convection scheme improves the warm biases in the deep Atlantic Ocean and Southern Ocean. This proof-of-concept study is a first step in developing the patchy parameterization scheme, which will be extended in future to use a prognostic eddy field as well as to parameterize convection due to under-ice brine rejection.

Highlights: • We compare the simulated Arctic Ocean in 15 global ocean–sea ice models. • There is a large spread in temperature bias in the Arctic Ocean between the models. • Warm bias models have a strong temperature anomaly of inflow of Atlantic Water. • Dense outflows formed on Arctic shelves are not captured accurately in the models. In this paper we compare the simulated Arctic Ocean in 15 global ocean-sea ice models in the framework of the Coordinated Ocean-ice Reference Experiments, phase II (CORE-II). Most of these models are the ocean and sea-ice components of the coupled climate models used in the Coupled Model Intercomparison Project Phase 5 (CMIP5) experiments. We mainly focus on the hydrography of the Arctic interior, the state of Atlantic Water layer and heat and volume transports at the gateways of the Davis Strait, the Bering Strait, the Fram Strait and the Barents Sea Opening. We found that there is a large spread in temperature in the Arctic Ocean between the models, and generally large differences compared to the observed temperature at intermediate depths. Warm bias models have a strong temperature anomaly of inflow of the Atlantic Water entering the Arctic Ocean through the Fram Strait. Another process that is not represented accurately in the CORE-II models is the formation of cold and dense water, originating on the eastern shelves. In the cold bias models, excessive cold water forms in the Barents Sea and spreads into the Arctic Ocean through the St. Anna Through. There is a large spread in the simulated mean heat and volume transports through the Fram Strait and the Barents Sea Opening. The models agree more on the decadal variability, to a large degree dictated by the common atmospheric forcing. We conclude that the CORE-II model study helps us to understand the crucial biases in the Arctic Ocean. The current coarse resolution state-of-the-art ocean models need to be improved in accurate representation of the Atlantic Water inflow into the Arctic and density currents coming from the shelves.

The Arctic Ocean simulated in fourteen global ocean-sea ice models in the framework of the Coordinated Ocean-ice Reference Experiments, phase II (CORE II) is analyzed. The focus is on the Arctic sea ice extent, the solid freshwater (FW) sources and solid freshwater content (FWC). Available observations are used for model evaluation. The variability of sea ice extent and solid FW budget is more consistently reproduced than their mean state in the models. The descending trend of September sea ice extent is well simulated in terms of the model ensemble mean. Models overestimating sea ice thickness tend to underestimate the descending trend of September sea ice extent. The models underestimate the observed sea ice thinning trend by a factor of two. When averaged on decadal time scales, the variation of Arctic solid FWC is contributed by those of both sea ice production and sea ice transport, which are out of phase in time. The solid FWC decreased in the recent decades, caused mainly by the reduction in sea ice thickness. The models did not simulate the acceleration of sea ice thickness decline, leading to an underestimation of solid FWC trend after 2000. The common model behavior, including the tendency to underestimate the trend of sea ice thickness and March sea ice extent, remains to be improved.

The Arctic Ocean has important freshwater sources including river runoff, low evaporation, and exchange with the Pacific Ocean. In the future, we expect even larger freshwater input as the global hydrological cycle accelerates, increasing high-latitude precipitation, and river runoff. Previous modeling studies show some robust responses to high-latitude freshwater perturbations, including a strengthening of Arctic stratification and a weakening of the large-scale ocean circulation; some idealized modeling studies also document a stronger cyclonic circulation within the Arctic Ocean itself. With the broad range of scales and processes involved, the overall effect of increasing runoff requires an understanding of both the local processes and the broader linkages between the Arctic and surrounding oceans. Here we adopt a more comprehensive modeling approach by increasing river runoff to the Arctic Ocean in a coupled ice-ocean general circulation model, and show contrasting responses in the polar and subpolar regions. Within the Arctic, the stratification strengthens, the halocline and Atlantic Water layer warm, and the cyclonic circulation spins up, in agreement with previous work. In the subpolar North Atlantic, the model simulates a colder and fresher water column with weaker barotropic circulation. In contrast to the estuarine circulation theory, the volume exchange between the Arctic Ocean and the surrounding oceans does not increase with increasing runoff. While these results are robust in our model, we require experiments with other model systems and more complete observational syntheses to better constrain the sensitivity of the climate system to high-latitude freshwater perturbations. publishedVersion

Simulated inter-annual to decadal variability and trends in the North Atlantic for the 1958-2007 period from twenty global ocean - sea-ice coupled models are presented. These simulations are performed as contributions to the second phase of the Coordinated Ocean-ice Reference Experiments (CORE-II). The study is Part II of our companion paper (Danabasoglu et al., 2014) which documented the mean states in the North Atlantic from the same models. A major focus of the present study is the representation of Atlantic meridional overturning circulation (AMOC) variability in the participating models. Relationships between AMOC variability and those of some other related variables, such as subpolar mixed layer depths, the North Atlantic Oscillation (NAO), and the Labrador Sea upper-ocean hydrographic properties, are also investigated. In general, AMOC variability shows three distinct stages. During the first stage that lasts until the mid-to late-1970s, AMOC is relatively steady, remaining lower than its long-term (1958-2007) mean. Thereafter, AMOC intensifies with maximum transports achieved in the mid-to late-1990s. This enhancement is then followed by a weakening trend until the end of our integration period. This sequence of low frequency AMOC variability is consistent with previous studies. Regarding strengthening of AMOC between about the mid-1970s and the mid-1990s, our results support a previously identified variability mechanism where AMOC intensification is connected to increased deep water formation in the subpolar North Atlantic, driven by NAO-related surface fluxes. The simulations tend to show general agreement in their temporal representations of, for example, AMOC, sea surface temperature (SST), and subpolar mixed layer depth variabilities. In particular, the observed variability of the North Atlantic SSTs is captured well by all models. These findings indicate that simulated variability and trends are primarily dictated by the atmospheric datasets which include the influence of ocean dynamics from nature superimposed onto anthropogenic effects. Despite these general agreements, there are many differences among the model solutions, particularly in the spatial structures of variability patterns. For example, the location of the maximum AMOC variability differs among the models between Northern and Southern Hemispheres. (C) 2015 Elsevier Ltd. All rights reserved.