Description: Highlights: • Phase II of the Coordinated Ocean-ice Reference Experiments (CORE-II) is introduced. • Solutions from CORE-II simulations from eighteen participating models are presented. • Mean states in the North Atlantic with a focus on AMOC are examined. • The North Atlantic solutions differ substantially among the models. • Many factors, including parameterization choices, contribute to these differences. Simulation characteristics from eighteen global ocean–sea-ice coupled models are presented with a focus on the mean Atlantic meridional overturning circulation (AMOC) and other related fields in the North Atlantic. These experiments use inter-annually varying atmospheric forcing data sets for the 60-year period from 1948 to 2007 and are performed as contributions to the second phase of the Coordinated Ocean-ice Reference Experiments (CORE-II). The protocol for conducting such CORE-II experiments is summarized. Despite using the same atmospheric forcing, the solutions show significant differences. As most models also differ from available observations, biases in the Labrador Sea region in upper-ocean potential temperature and salinity distributions, mixed layer depths, and sea-ice cover are identified as contributors to differences in AMOC. These differences in the solutions do not suggest an obvious grouping of the models based on their ocean model lineage, their vertical coordinate representations, or surface salinity restoring strengths. Thus, the solution differences among the models are attributed primarily to use of different subgrid scale parameterizations and parameter choices as well as to differences in vertical and horizontal grid resolutions in the ocean models. Use of a wide variety of sea-ice models with diverse snow and sea-ice albedo treatments also contributes to these differences. Based on the diagnostics considered, the majority of the models appear suitable for use in studies involving the North Atlantic, but some models require dedicated development effort.

Description: Highlights: • Global mean sea level simulated in interannual CORE simulations. • Regional sea level patterns simulated in interannual CORE simulations. • Theoretical foundation for analysis of global mean sea level and regional patterns. Abstract: We provide an assessment of sea level simulated in a suite of global ocean-sea ice models using the interannual CORE atmospheric state to determine surface ocean boundary buoyancy and momentum fluxes. These CORE-II simulations are compared amongst themselves as well as to observation-based estimates. We focus on the final 15 years of the simulations (1993–2007), as this is a period where the CORE-II atmospheric state is well sampled, and it allows us to compare sea level related fields to both satellite and in situ analyses. The ensemble mean of the CORE-II simulations broadly agree with various global and regional observation-based analyses during this period, though with the global mean thermosteric sea level rise biased low relative to observation-based analyses. The simulations reveal a positive trend in dynamic sea level in the west Pacific and negative trend in the east, with this trend arising from wind shifts and regional changes in upper 700 m ocean heat content. The models also exhibit a thermosteric sea level rise in the subpolar North Atlantic associated with a transition around 1995/1996 of the North Atlantic Oscillation to its negative phase, and the advection of warm subtropical waters into the subpolar gyre. Sea level trends are predominantly associated with steric trends, with thermosteric effects generally far larger than halosteric effects, except in the Arctic and North Atlantic. There is a general anti-correlation between thermosteric and halosteric effects for much of the World Ocean, associated with density compensated changes.

Description: We estimate 3-dimensional ocean currents from hydrographic (ARGO) data by determining the associated circulation in an inverse model. While velocities are treated diagnostic as an instantaneous steady state response temperature and salinity are allowed to change slowly with their inter-annual variability. Annual mean solutions are presented for 1999 to 2008. Altimetry referenced to a geoid provides a mean dynamic topography that determines the large scale surface circulation. AGRO data extend this information further into the ocean. It appeared useful to regularize the solution by constraining deep velocities to be small or as in our case to be close to a prognostic model simulation. Altimetry alone already improves temperature and salinity fields while ARGO data are less useful in constraining the dynamic topography. Heat, volume and overturning transports are in general agreement with previous work. Their inter-annual variability appears to be large in comparison to possible trends. Transport variances are estimated by perturbing the input data in a Monte Carlo simulation. They are smaller than the changes between consecutive years. ARGO data coverage seems to be reliable for this work after the year 2002.

Description: A new climate model supporting multi-resolution meshes in the ocean component has been established at the Alfred Wegener Institute (AWI) in Bremerhaven. The atmospheric component is ECHAM6 with T63L47 setting, while the ocean is simulated by the AWI multi-resolution model FESOM, supporting triangular unstructured meshes. Two multi-century simulations with ECHAM6-FESOM, REF and TRO, document the beneficial role of an increased tropical ocean resolution for ENSO simulations. REF features a tropical ocean resolution of about 1°, TRO employs more than 0.25° in a narrow equatorial band, with resolution gradually decreasing to 1° as in REF. Outside the tropical belt (15°N to 15°S), both meshes are identical. REF and TRO simulate a mean climate comparable to some of the best CMIP5 models. In TRO, however, both the cold tongue SST bias and the western Pacific SST standard deviation bias appear to improve along with the Nino-3 index statistics. Also, advanced ENSO diagnostics including the Nino-3.4 seasonal variance, the annual cycle representation, and its interaction with ENSO tend to improve. The robustness of these improvements is analyzed and their physical explanations are explored.

Description: We analyse the ENSO-like variability in the newly established global climate model ECHAM-FESOM. This is the first global coupled model with an ocean module supporting unstructured meshes. The Finite Element Sea Ice - Ocean Model (FESOM) is a dynamical ocean model development at AWI Bremerhaven. In contrast to conventional ocean models, the spatial discretization is based on the Finite Element method. This method allows a variable spatial resolution of the triangular surface mesh with high mesh-stretching factors. FESOM has been used in numerous recent, yet uncoupled, studies. Its validation in the climate context is still ongoing activity. ECHAM is a state-of-the-art spectral atmosphere model developed at the Max-Planck-Institute for Meteorology in Hamburg for climate modelling purposes. We apply the latest generation, version 6, with a T63L47 resolution. FESOM and ECHAM are currently coupled by the OASIS3-MCT coupler and a structured exchange mesh. We analyse two simulation runs that differ in the tropical ocean mesh resolution between 15°N and 15°S. Setup 1 uses a reference mesh with a resolution of about 1° in the tropics. In contrast, Setup 2 has a higher resolution of 1/4° (in a narrow band around the equator) that gradually decreases to 1°. Outside the tropics both meshes are identical. Modelled Nino3.4 indices are compared with observations and the influence of the mesh resolution is discussed.

Description: A review is given of existing efforts and future challenges in large-scale ocean modeling on unstructured meshes. Because of large integration time the large-scale ocean circulation models require more attention with respect to conservation and accuracy than their coastal counterparts, and deal with different dynamics. Numerous discretizations of finite-element and finite-volume type have been proposed and explored, but only simplest approaches (like P1-P1 of FESOM or cell-vertex of FVCOM) are realized as working tools. While the results of first applications performed in global context are very encouraging and show the feasibility of local refinement by a factor of 30-50 in a global setup, they also indicate that further efforts are needed if one aims at simulating the ocean in eddying regimes. These efforts must concentrate on numerical efficiency of unstructured-mesh codes and ensure much higher accuracy of advection schemes than currently available on linear elements. Since the selection of discretization introduces obvious limitations on unstructured meshes, one has to re-consider it from the perspective that takes into account the already available lessons, and not only the representation of linear wave dynamics in the shallow-water context. In particular, too large velocity spaces of many mixed discretizations as a rule lead to difficulties in the performance of momentum advection in eddying regimes, and continuous representation of scalar quantities creates many inconveniences in hydrostatic codes. While the need for improved advection schemes tells on its own in favor of either discontinuous Galerkin methods or high-order reconstructions with finite volumes, the key question is how to implement them without significant loss of efficiency. The area of large-scale ocean simulations is almost fully dominated by structured-mesh models which are currently much more accurate and efficient per degree of freedom than models on unstructured meshes. The acceptance of unstructured-mesh technology on this `large-scale background' depends on our progress with this question as well as ability to show on practical examples that refinement is more practical and consistent than nesting.

Description: The data available from satellite altimetry and Argo profiling buoys are combined into Inverse Finite Element Ocean model (IFEOM) for each of the years 1999 and 2008. The model solves for temperature and salinity fields that are close to measurements, respects stationary dynamical balances, and simultaneously produces estimates of the circulation. Several experiments have been performed to reconstruct the interannual variability. We show, that including altimetry improves the circulation picture. We estimate variability of transports and heat content as result of combined assimilation of Argo and altimetry.