• AMAP, 1998: AMAP assessment report: Arctic pollution issues. Arctic Monitoring and Assessment Programme, 859 pp.

  • Antonov, J. I., , Locarnini R. A. , , Boyer T. P. , , Mishonov A. V. , , and Garcia H. E. , 2006: Salinity. Vol. 2, World Ocean Atlas 2005, NOAA Atlas NESDIS 62, 182 pp.

  • Arakawa, A., , and Lamb V. R. , 1977: Computational design of the basic dynamical processes of the UCLA general circulation model. Methods in Computational Physics, J. Chang, Ed., Vol. 17, Academic Press, 173–265.

  • Beszczynska-Moeller, A., , Woodgate R. A. , , Lee C. , , Melling H. , , and Karcher M. , 2011: A synthesis of exchanges through the main oceanic gateways to the Arctic Ocean. Oceanography, 24, 8299.

    • Search Google Scholar
    • Export Citation
  • Böning, C. W., , Scheinert M. , , Dengg J. , , Biastoch A. , , and Funk A. , 2006: Decadal variability of subpolar gyre transport and its reverberation in the North Atlantic overturning. Geophys. Res. Lett., 33, L21S01, doi:10.1029/2006GL026906.

    • Search Google Scholar
    • Export Citation
  • Bryan, K., , Manabe S. , , and Pacanowski R. C. , 1975: A global ocean–atmosphere climate model. Part II. The oceanic circulation. J. Phys. Oceanogr., 5, 3046.

    • Search Google Scholar
    • Export Citation
  • Canuto, V. M., , Howard A. , , Cheng Y. , , and Dubovikov M. S. , 2001: Ocean turbulence. Part I: One-point closure model—Momentum and heat vertical diffusivities. J. Phys. Oceanogr., 31, 14131426.

    • Search Google Scholar
    • Export Citation
  • Canuto, V. M., , Howard A. , , Cheng Y. , , and Dubovikov M. S. , 2002: Ocean turbulence. Part II: Vertical diffusivities of momentum, heat, salt, mass, and passive scalars. J. Phys. Oceanogr., 32, 240264.

    • Search Google Scholar
    • Export Citation
  • Canuto, V. M., , Howard A. , , Cheng Y. , , Muller, C. J. , Leboissetier A. , , and Jayne S. R. , 2010: Ocean turbulence, III: New GISS vertical mixing scheme. Ocean Modell., 34, 7091, doi:10.1016/j.ocemod.2010.04.006.

    • Search Google Scholar
    • Export Citation
  • Coachman, L. K., , and Barnes C. A. , 1961: The contribution of Bering Sea water to the Arctic Ocean. Arctic, 14, 147161.

  • Courant, R., , Friedrichs K. , , and Lewy H. , 1928: Über die partiellen Differenzengleichungen der mathematischen Physik. Math. Ann., 100, 3274.

    • Search Google Scholar
    • Export Citation
  • Coward, A. C., , Killworth P. D. , , and Blundell J. R. , 1994: Tests of a two-grid world ocean model. J. Geophys. Res., 99 (C11), 22 72522 735.

    • Search Google Scholar
    • Export Citation
  • Danabasoglu, G., , Bates S. C. , , Briegleb B. P. , , Jayne S. R. , , Jochum M. , , Large W. G. , , Peacock S. , , and Yeager S. G. , 2012: The CCSM4 ocean component. J. Climate, 25, 13611389.

    • Search Google Scholar
    • Export Citation
  • DiNezio, P. N., , Gramer L. J. , , Johns W. E. , , Meinen C. S. , , and Baringer M. O. , 2009: Observed interannual variability of the Florida Current: Wind forcing and the North Atlantic Oscillation. J. Phys. Oceanogr., 39, 721736.

    • Search Google Scholar
    • Export Citation
  • Eby, M., , and Holloway G. , 1994: Grid transformation for incorporating the Arctic in a global ocean model. Climate Dyn., 10, 241247.

  • Flato, G. M., , Boer G. J. , , Lee W. G. , , McFarlane N. A. , , Ramsden D. , , Reader M. C. , , and Weaver A. J. , 2000: The Canadian Centre for Climate Modelling and Analysis global coupled model and its climate. Climate Dyn., 16, 451467.

    • Search Google Scholar
    • Export Citation
  • Furevik, T., , Bentsen M. , , Drange H. , , Kindem I. K. T. , , Kvamstø N. G. , , and Sorteberg A. , 2003: Description and evaluation of the Bergen climate model: ARPEGE coupled with MICOM. Climate Dyn., 21, 2751.

    • Search Google Scholar
    • Export Citation
  • Gammelsrød, T., , Leikvin Ø. , , Lien V. , , Budgell W. P. , , Loeng H. , , and Maslowski W. , 2009: Mass and heat transports in the NE Barents Sea: Observations and models. J. Mar. Syst., 75, 5669.

    • Search Google Scholar
    • Export Citation
  • Gao, Y., , and Yu L. , 2008: Subpolar gyre index and the North Atlantic meridional overturning circulation in a coupled climate model. Atmos. Oceanic Sci. Lett., 1, 2932.

    • Search Google Scholar
    • Export Citation
  • Gent, P. R., , Bryan F. O. , , Danabasoglu G. , , Doney S. C. , , Holland W. R. , , Large W. G. , , and McWilliams J. C. , 1998: The NCAR Climate System Model global ocean component. J. Climate, 11, 12871306.

    • Search Google Scholar
    • Export Citation
  • Gordon, C., , Cooper C. , , Senior C. A. , , Banks H. , , Gregory J. M. , , Johns T. C. , , Mitchell J. F. B. , , and Wood R. A. , 2000: The simulation of SST, sea ice extents and ocean heat transports in a version of the Hadley Centre coupled model without flux adjustments. Climate Dyn., 16, 147168.

    • Search Google Scholar
    • Export Citation
  • Griffies, S. M., and Coauthors, 2000: Developments in ocean climate modelling. Ocean Modell., 2, 123192.

  • Griffies, S. M., and Coauthors, 2005: Formulation of an ocean model for global climate simulations. Ocean Sci., 1, 4579.

  • Häkkinen, S., , and Rhines P. B. , 2004: Decline of subpolar North Atlantic circulation during the 1990s. Science, 304, 555559.

  • Haney, R. L., 1971: Surface thermal boundary condition for ocean circulation models. J. Phys. Oceanogr., 1, 241248.

  • Huang, W.-Y., , Wang B. , , Yu Y.-Q. , , and Li L.-J. , 2014: Improvements in LICOM2. Part I: Vertical mixing. J. Atmos. Oceanic Technol., in press.

  • Kara, A. B., , Hurlburt H. E. , , and Wallcraft A. J. , 2005: Stability-dependent exchange coefficients for air–sea fluxes. J. Atmos. Oceanic Technol., 22, 10801094.

    • Search Google Scholar
    • Export Citation
  • Li, L. J., and Coauthors, 2013: The Flexible Global Ocean-Atmosphere-Land System Model, version g2: FGOALS-g2. Adv. Atmos. Sci., 543–560, doi:10.1007/s00376-012-2140-6.

    • Search Google Scholar
    • Export Citation
  • Liu, H.-L., , Lin P.-F. , , Yu Y.-Q. , , and Zhang X.-H. , 2012: The baseline evaluation of LASG/IAP climate system ocean model (LICOM) version 2. Acta Meteor. Sin., 26, 318329.

    • Search Google Scholar
    • Export Citation
  • Liu, X.-Y., , Liu H.-L. , , Zhang X.-H. , , and Yu R.-C. , 2006: Experiments of a reduced grid in LASG/IAP World Ocean general circulation models (OGCMs). J. Trop. Meteor., 12, 915.

    • Search Google Scholar
    • Export Citation
  • Locarnini, R. A., , Mishonov A. V. , , Antonov J. I. , , Boyer T. P. , , and Garcia H. E. , 2006: Temperature. Vol. 1, World Ocean Atlas 2005, NOAA Atlas NESDIS 61, 182 pp.

  • Lumpkin, R., , and Speer K. , 2007: Global ocean meridional overturning. J. Phys. Oceanogr., 37, 25502562.

  • Madec, G., 2008: NEMO ocean engine. IPSL Note du Pole de modélisation 27, 357 pp.

  • Mesinger, F., 1984: A blocking technique for representation of mountains in atmospheric models. Riv. Meteor. Aeronaut., 44, 195202.

  • Mesinger, F., , Janjić Z. I. , , Ničković S. , , Gavrilov D. , , and Deaven D. G. , 1988: The step-mountain coordinate: Model description and performance for cases of Alpine lee cyclogenesis and for a case of an Appalachian redevelopment. Mon. Wea. Rev., 116, 14931518.

    • Search Google Scholar
    • Export Citation
  • Murray, R. J., 1996: Explicit generation of orthogonal grids for ocean models. J. Comput. Phys., 126, 251273.

  • Murray, R. J., , and Reason C. J. C. , 2001: A curvilinear version of the Bryan–Cox–Semtner ocean model and its representation of the Arctic circulation. J. Comput. Phys., 171, 146.

    • Search Google Scholar
    • Export Citation
  • Murray, R. J., , and Reason C. J. C. , 2002: Fourier filtering and coefficient tapering at the North Pole in OGCMs. Ocean Modell., 4, 125.

    • Search Google Scholar
    • Export Citation
  • Pacanowski, R. C., , Dixon K. W. , , and Rosati A. , 1993: The GFDL Modular Ocean Model users guide. Geophysical Fluid Dynamics Laboratory Ocean Group Tech. Rep. 2, 46 pp.

  • Rudels, B., , Marnela M. , , and Eriksson P. , 2008: Constraints on estimating mass, heat and freshwater transports in the Arctic Ocean: An exercise. Arctic–Subarctic Ocean Fluxes, R. Dickson, J. Meincke, and P. Rhines, Eds., Springer Netherlands, 315–341.

  • Simmons, A. J., , and Gibson J. K. , 2000: The ERA-40 project plan. ERA-40 Project Report Series 1, ECMWF, 63 pp.

  • Skagseth, Ø., , Furevik T. , , Ingvaldsen R. , , Loeng H. , , Mork K. , , Orvik K. , , and Ozhigin V. , 2008: Volume and heat transports to the Arctic Ocean via the Norwegian and Barents Seas. Arctic–Subarctic Ocean Fluxes, R. Dickson, J. Meincke, and P. Rhines, Eds., Springer Netherlands, 45–64.

  • Smedsrud, L. H., , Ingvaldsen R. , , Nilsen J. E. Ø. , , and Skagseth Ø. , 2010: Heat in the Barents Sea: Transport, storage, and surface fluxes. Ocean Sci., 6, 219234.

    • Search Google Scholar
    • Export Citation
  • Smith, R., and Coauthors, 2010: The Parallel Ocean Program (POP) reference manual: Ocean component of the Community Climate System Model (CCSM). Los Alamos National Laboratory Tech. Rep. LAUR-10-01853, 141 pp.

  • Steele, M., , Morley R. , , and Ermold W. , 2001: PHC: A global ocean hydrography with a high-quality Arctic Ocean. J. Climate, 14, 20792087.

    • Search Google Scholar
    • Export Citation
  • Sun, H.-C., , and Zhou G.-Q. , 2010: An Arctic sea ice simulation using an ocean-ice coupled model. Atmos. Oceanic Sci. Lett., 3, 219223.

    • Search Google Scholar
    • Export Citation
  • Woodgate, R. A., , Aagaard K. , , and Weingartner T. J. , 2005: Monthly temperature, salinity, and transport variability of the Bering Strait through flow. Geophys. Res. Lett., 32, L04601, doi:10.1029/2004GL021880.

    • Search Google Scholar
    • Export Citation
  • Xu, S., and Coauthors, 2013: Simulation of sea ice in FGOALS-g2: Climatology and late 20th century changes. Adv. Atmos. Sci., 30, 658673.

    • Search Google Scholar
    • Export Citation
  • Yu, R., 1989: The design of a regional numerical forecasting model with steep topography. Sci. Atmos. Sin., 13, 139149.

  • Yu, R., 1994: A test for numerical weather prediction of real-time for China flood season precipitation in 1993 by a regional eta-coordinate model. Sci. Atmos. Sin., 18, 284292.

    • Search Google Scholar
    • Export Citation
  • Yu, Y.-Q., 1997: Design of ocean-sea ice-atmosphere coupling scheme and numerical simulation of interdecadal climate variability. Institute of Atmospheric Physics, Chinese Academy of Sciences, 130 pp.

  • View in gallery

    Land–sea distribution and some critical straits/channels in the Arctic Ocean. The regions with gray and white colors correspond to land and ocean, respectively. The colored lines associated with red letters indicate the critical straits/channels, across which the net volume fluxes are calculated. Section “Bering” is for the Bering Strait, section W is between 83° and 90°N along 85.5°W, section E is between 82° and 90°N along 94.5°E, section “Fram” is for the Fram Strait between 17°W and 11°E along 79.5°N, section BSO is between 69° and 77°N along 16.5°E, and section BSX is between 77° and 80°N along 61°E.

  • View in gallery

    Time series of annual-mean, global area-weighted average of SST bias (°C) and SSS bias (psu) from Exps IMPV0 (blue curve) and IMPV1 (red curve) during years 1–650. The observations used for bias estimations are from the PHC3.0 dataset.

  • View in gallery

    (left) Annual-mean SST bias (°C) and (right) SSS bias (psu) in the Arctic Ocean of (top) Exps IMPV0 and (middle) IMPV1. The observations used for bias estimations are from the PHC3.0 dataset. (bottom) Differences between the two experiments, using finer contour intervals. The model output is taken from years 601–650.

  • View in gallery

    Annual-mean surface currents in the Arctic Ocean. This figure is a reproduction of Fig. 3.29 of AMAP (AMAP 1998).

  • View in gallery

    Arctic Ocean currents in (left) winter, (middle) summer, and (right) its annual mean at 15-m depth from (top) Exps IMPV0 and (bottom) IMPV1. The arrows indicate the directions of currents, while the colors give their velocities (m s−1). The model output is taken from years 601–650.

  • View in gallery

    Mean annual cycles of the net volume fluxes in IMPV0 (blue curves) and IMPV1 (red curves) across (a) the region 86°–90°N across sections W and E, and (b) sections W and E. The units are in Sv. The positive sign is used for eastward volume flux across section W, and for the westward volume flux across section E. The model output is taken from years 601–650.

  • View in gallery

    Mean annual cycles of net volume fluxes (Sv) in IMPV0 (blue curves) and IMPV1 (red curves) across the (a) Bering Strait, (b) Fram Strait, (c) BSO, and (d) BSX. The positive sign is used for the eastward/northward volume flux. The model output is taken from years 601–650.

  • View in gallery

    Annual-mean, zonally integrated overturning streamfunction (Sv) in (left) the global ocean and (right) the Atlantic Ocean from (top) Exp IMPV0 and (middle) Exp IMPV1. (bottom) Differences between the two experiments. The model output is taken from model years 601–650.

  • View in gallery

    (left) Annual-mean barotropic streamfunction (Sv) and (right) mixed layer depth (m) in the Atlantic Ocean from (top) Exp IMPV0 and (middle) Exp IMPV1. The differences between the two experiments are shown in the bottom. The model output is taken from model year 601–650.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 19 19 0
PDF Downloads 14 14 3

Improvements in LICOM2. Part II: Arctic Circulation

View More View Less
  • 1 Ministry of Education Key Laboratory for Earth System Modeling, and Center for Earth System Science, Tsinghua University, Beijing, China
  • | 2 Ministry of Education Key Laboratory for Earth System Modeling, and Center for Earth System Science, Tsinghua University, and State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
  • | 3 State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China
© Get Permissions
Full access

Abstract

A known issue of the National Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics/Institute of Atmospheric Physics Climate Ocean Model, version 2 (LICOM2, the standard version) is the use of an artificial island in the Arctic Ocean. The computational instability in the polar region seriously influences the model performance in terms of the Arctic circulation. The above-mentioned instability was originally attributed to the converging zonal grids in the polar region. However, this study finds that better computational stability could be achieved in an improved version of LICOM2 (i.e., LICOM2_imp) after four improvements on implementations of the vertical mixing, mesoscale eddy parameterization, and bottom drag schemes. LICOM2_imp is then able to reduce the aforesaid artificial island to a point (i.e., the North Pole).

Two experiments of 650-yr integration by LICOM2_imp are carried out using different bathymetries: Exp IMPV0 with the artificial island (88°–90°N) and IMPV1 with only the single pole. The focus of this paper is on the Arctic circulation. Exp IMPV1 gives a more reasonable distribution of salinity and temperature in the Arctic Ocean, a more accurate location of the center of the Beaufort Gyre, and a better net volume flux of the transpolar drift. With more realistic bathymetry in the Arctic Ocean, the biases of net volume fluxes across the Fram Strait, Barents Sea Opening, and Barents Sea Exit are reduced from 1.71 to 1.56, from 0.23 to 0.10, and from 0.71 to 0.45 Sv (1 Sv ≡ 106 m3 s−1), respectively, closer to the observations. The large biases of the net volume fluxes at the Fram Strait in both experiments may be attributed to the closed Nares Strait and other straits/channels in the Canadian Arctic Archipelago.

Corresponding author address: Bin Wang, Room 815, Mengminwei Building, Tsinghua University, Haidian District, Beijing 100084, China. E-mail: wab@mail.tsinghua.edu.cn

Abstract

A known issue of the National Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics/Institute of Atmospheric Physics Climate Ocean Model, version 2 (LICOM2, the standard version) is the use of an artificial island in the Arctic Ocean. The computational instability in the polar region seriously influences the model performance in terms of the Arctic circulation. The above-mentioned instability was originally attributed to the converging zonal grids in the polar region. However, this study finds that better computational stability could be achieved in an improved version of LICOM2 (i.e., LICOM2_imp) after four improvements on implementations of the vertical mixing, mesoscale eddy parameterization, and bottom drag schemes. LICOM2_imp is then able to reduce the aforesaid artificial island to a point (i.e., the North Pole).

Two experiments of 650-yr integration by LICOM2_imp are carried out using different bathymetries: Exp IMPV0 with the artificial island (88°–90°N) and IMPV1 with only the single pole. The focus of this paper is on the Arctic circulation. Exp IMPV1 gives a more reasonable distribution of salinity and temperature in the Arctic Ocean, a more accurate location of the center of the Beaufort Gyre, and a better net volume flux of the transpolar drift. With more realistic bathymetry in the Arctic Ocean, the biases of net volume fluxes across the Fram Strait, Barents Sea Opening, and Barents Sea Exit are reduced from 1.71 to 1.56, from 0.23 to 0.10, and from 0.71 to 0.45 Sv (1 Sv ≡ 106 m3 s−1), respectively, closer to the observations. The large biases of the net volume fluxes at the Fram Strait in both experiments may be attributed to the closed Nares Strait and other straits/channels in the Canadian Arctic Archipelago.

Corresponding author address: Bin Wang, Room 815, Mengminwei Building, Tsinghua University, Haidian District, Beijing 100084, China. E-mail: wab@mail.tsinghua.edu.cn

1. Introduction

According to the Courant–Friedrichs–Lewy (CFL) condition (Courant et al. 1928), the time step for the finite difference algorithm should satisfy
e1
where is the time step, is the grid spacing, and is the velocity of the fastest wave, such as gravity wave. For a longitude/latitude grid, the convergence of the zonal circle at Earth’s pole leads to polar singularity—that is, the grid spacing and the corresponding time step approach zero (Griffies et al. 2000) and the direction of the current at the pole cannot be determined. The difficulty in determining the direction of the current at the pole can be overcome by using a staggered grid with the scalar quantity, such as temperature/salinity, being placed at the North Pole. To avoid the grid spacing related to the polar singularity in the longitude/latitude grid, there are several choices: 1) using an artificial island that covers the North Pole and several surrounding latitudinal circles (Pacanowski et al. 1993; Flato et al. 2000), 2) using a numerical filtering that is applied to the high latitudes (Bryan et al. 1975; Murray and Reason 2002), and 3) using both artificial island and numerical filtering. Because of its advantages in simulating the poleward heat transport induced by differential solar heating and characterizing the rotation of the earth, a longitude/latitude grid is still used by three global ocean models of the coupled climate models that take part in the phase 5 of the Coupled Model Intercomparison Project (CMIP5)—that is, the National Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics (LASG)/Institute of Atmospheric Physics (IAP) Climate Ocean Model, version 2 (LICOM2; Liu et al. 2012); the ocean component of the Hadley Centre Coupled Model (Gordon et al. 2000); and the ocean component of the Canadian Earth System Model (Gent et al. 1998). In each of these three longitude/latitude ocean models, numerical filtering is used; in addition, the LICOM2 uses an artificial island in the Arctic Ocean from 88° to 90°N. Currently, grid transformation techniques are widely used in global ocean models for treating the polar singularity or for including the whole Arctic Ocean: 1) a combination of two coordinate systems (Coward et al. 1994; Eby and Holloway 1994)—that is, two grid systems are patched together with a suitable boundary; 2) a dipole or tripole grid system (Murray 1996; Murray and Reason 2001)—that is, a grid with two or three “poles” located over landmasses, which are widely used, including the Modular Ocean Model, version 4 (MOM4; Griffies et al. 2005); the Parallel Ocean Program, version 2 (POP2; Smith et al. 2010; Danabasoglu et al. 2012); and the Nucleus for European Modelling of the Ocean, version 3.4 (NEMO3.4; Madec 2008). In addition, the tripole grid system can be viewed as a combination of a longitude/latitude grid and a dipole grid.

Basic characteristics of the World Ocean circulation are well captured by the standard version of LICOM2 (hereafter LICOM2; Liu et al. 2012). However, weaknesses have been reported due to the use of the artificial island in the Arctic Ocean in LICOM2. Sun and Zhou (2010) found that the simulated Arctic sea ice drifted around—rather than across—the North Pole as observed, when LICOM2 was coupled to the sea ice model of the Bergen Climate Model (Furevik et al. 2003). In the analysis of the twentieth-century simulation results of the Flexible Global Ocean–Atmosphere–Land System Model gridpoint, version 2.0 (FGOALS-g2; Li et al. 2013) that takes LICOM2 as its ocean component, Xu et al. (2013) found that the Arctic sea ice is much thicker than the observation and the transpolar drift has to go around the artificial island. These distortions are clearly caused by the artificial island used in the ocean model.

Efforts have been made to reduce the distortions caused by the artificial island in LICOM2 and its predecessors. Yu (1997) included all the latitudes except the North Pole in an early version of the LASG/IAP ocean model, a predecessor of LICOM2, with a lower horizontal resolution of 4° × 5°. By reducing the zonal grid numbers toward the North Pole in a nominal 1° resolution version of LICOM2, Liu et al. (2006) extended the model to 89°N and left only the North Pole as the artificial island. However, because of the difficulty in implementing some parameterization schemes (e.g., mesoscale eddy parameterization) in a reduced zonal grid (Liu et al. 2006), LICOM2 has to use the unreduced longitude/latitude grid and thus it cannot include all the latitudes besides the North Pole as in Yu (1997) and Liu et al. (2006); otherwise, it becomes computationally unstable, which has commonly been attributed to the numerical scheme.

This paper is Part II of a study that focuses on the performance of an improved version of LICOM2 (named “LICOM2_imp”). In Huang et al. (2014, hereafter Part I), efforts have been made to improve the long-term numerical stability of LICOM2, through the correction of vertical interpolation related to the vertical mixing scheme, the replacement of the nonconvergent iteration method that solves the vertical mixing scheme by an analytic method, the correction of unmatched mesoscale eddy parameterized grid, and the removal of the deflection effect on the bottom drag by the Coriolis force. Better numerical stability of LICOM2_imp allows a significant reduction of the artificial island.

Part I of this study evaluated the basic performance of LICOM2_imp (Huang et al. 2014). With the same artificial island (88°–90°N) in the Arctic Ocean as in LICOM2, LICOM2_imp outperforms LICOM2 in simulating the sea surface temperature (SST), sea surface salinity (SSS), the Atlantic meridional overturning circulation (AMOC), and the Antarctic Circumpolar Current. This study is mainly devoted to evaluating the performance of LICOM2_imp in simulating the Arctic circulation.

The organization of the paper is as follows. In section 2, experiment design and datasets used for initializing, forcing, and validating the ocean model are detailed. In section 3, the Arctic circulation of LICOM2_imp is evaluated, and the influence of the size of artificial island is also discussed. Summary and conclusions, as well as some suggestions for further model development, are given in section 4.

2. Model description, experiment design, and datasets used

a. Experiment design

Liu et al. (2012) provided detailed descriptions of LICOM2. Huang et al. (2014) introduced LICOM2_imp and described the differences of scientific design and numerical implementations between LICOM2 and LICOM2_imp. Therefore, we will only briefly describe LICOM2_imp. In the horizontal direction, the longitude/latitude Arakawa B grid (Arakawa and Lamb 1977) is adopted. The nominal horizontal resolution is 1°, with a refinement to 0.5° in the meridional direction in the equatorial region. To obtain the horizontal pressure gradient with small bias, the eta coordinate with a stepwise bottom boundary is employed along the vertical direction (Mesinger 1984; Mesinger et al. 1988; Yu 1989, 1994). There are 30 vertical layers in the vertical direction with 15 layers in the upper 150 m. A process-splitting-based algorithm is used for time integration. The vertical mixing scheme is the second-order turbulence model from Canuto et al. (2001, 2002, 2010). A Haney-type surface thermal boundary condition (Haney 1971) is taken for the SST, while the coupling coefficients are from the Ocean Model Intercomparison Project forcing (Simmons and Gibson 2000; Kara et al. 2005). The restoring boundary condition is used for the SSS with a time scale of 30 days. The ocean model does not include a sea ice component, a simple treatment is taken—that is, −1.8°C is taken as the ice point and all the temperatures lower than it are set to this ice point.

To evaluate the simulation of the Arctic circulation by LICOM2_imp and the influence of the size of the artificial island on the simulation of Arctic circulation, two experiments are carried out in this paper. The first experiment, named Exp IMPV0, which is obtained with the same artificial island used in the standard version of LICOM2—that is, the model domain is 78°S–87°N. Because the improvements led to a much better computational stability in LICOM2_imp (Huang et al. 2014), we are able to reduce the size of the artificial island; therefore, Exp IMPV1 is the same as Exp IMPV0, except with the smallest artificial island (i.e., with the North Pole only). The maximum depth of the whole model domain is 5600 m, while that in the Arctic Ocean is about 4200 m. The land–sea distribution with some important straits/channels marked in/near the Arctic Ocean of Exp IMPV1 is shown in Fig. 1. The Nares Strait and the straits/channels in the Canadian Arctic Archipelago are closed due to the horizontal resolution is 1° there. Both experiments are initialized from the same climate mean state of potential temperature and salinity with no motion and integrated for 650 years. In the latitudes higher than 65°, polar filtering is applied to the barotropic modes once every day and to the baroclinic modes once every 4 days. For the settings of time steps, mesoscale eddy parameterization, and horizontal viscosity, readers are referred to Huang et al. (2014).

Fig. 1.
Fig. 1.

Land–sea distribution and some critical straits/channels in the Arctic Ocean. The regions with gray and white colors correspond to land and ocean, respectively. The colored lines associated with red letters indicate the critical straits/channels, across which the net volume fluxes are calculated. Section “Bering” is for the Bering Strait, section W is between 83° and 90°N along 85.5°W, section E is between 82° and 90°N along 94.5°E, section “Fram” is for the Fram Strait between 17°W and 11°E along 79.5°N, section BSO is between 69° and 77°N along 16.5°E, and section BSX is between 77° and 80°N along 61°E.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

b. Datasets

The potential temperature and salinity used for initializing the ocean model are from the World Ocean Atlas 2005 (WOA05; Locarnini et al. 2006; Antonov et al. 2006). The forcing data used at the sea surface are the long-term-mean annual cycle from the 24-h forecast of the 40-yr European Centre for Medium-Range Weather Forecasts Re-Analysis (ERA-40; Simmons and Gibson 2000; Kara et al. 2005) of 1979–93, which include wind stresses, shortwave radiation, nonsolar heat fluxes, and coupling coefficients. All these input forcings are in a 12-month format and interpolated linearly to 365-day format for the ocean model. The data used for validation include the ocean temperature and salinity data from the Polar Science Center Hydrographic Climatology, version 3.0 (PHC3.0; Steele et al. 2001); the estimated decadal mean of AMOC streamfunction by inverse techniques based on air–sea fluxes of heat and freshwater, hydrographic sections, and direct current measurements (Lumpkin and Speer 2007); and the surface currents in the Arctic Ocean from the Arctic Monitoring and Assessment Programme (AMAP; AMAP 1998). In addition, to give reference values for the net volume fluxes across the straits/channels in the Arctic Ocean (Fig. 1), the observed net volume flux for the Bering Strait from 1990 to 2004 by moored measurements (Woodgate et al. 2005), the net volume flux for the Fram Strait estimated by hydrographic sections from 1980 to 2005 (Rudels et al. 2008; Beszczynska-Moeller et al. 2011), the observed net volume flux across the Barents Sea Opening (BSO) from 1997 to 2007 by current meter and hydrographic data (Smedsrud et al. 2010; Skagseth et al. 2008; Beszczynska-Moeller et al. 2011), and the observed net volume flux across the Barents Sea Exit (BSX; Gammelsrød et al. 2009) are used.

3. Evaluation of Arctic circulation in LICOM2_imp

The performance of LICOM2_imp in the Arctic Ocean is evaluated in terms of the following variables and features: potential temperature, salinity, currents, and net volume fluxes through some important straits/channels. In addition, the influence of the size of artificial island on important current system—that is, the AMOC—is also discussed. In general, the use of a smaller artificial island in Exp IMPV1 leads to improvements over Exp IMPV0 in the Arctic Ocean and at some adjacent straits/channels, and also to slight improvements in simulating the maximum of AMOC streamfunction.

a. Potential temperature and salinity

Both experiments show good stability in their 650-yr integrations, and the global area-weighted mean SST and SSS approach their steady states (Fig. 2): the long-term trends of SST are 1.9 × 10−5 and 2.2 × 10−5 °C yr−1 and those of SSS are 5.8 × 10−6 and 6.9 × 10−6 psu yr−1 for IMPV0 and IMPV1, respectively. Comparing with the observed global-mean SST and SSS from PHC3.0 (Steele et al. 2001) over the period from 1979 to 1993 [which are 18.19°C and 34.73 psu, respectively; standard deviations (SDs) are not available]; however, both IMPV0 and IMPV1 overestimate the global-mean SST while underestimating the global-mean SSS. The reduced artificial island contributes to a colder SST and lower SSS simulations in the “added ocean region” of IMPV1, which results in an alleviated warm bias of the global-mean SST but a slightly enlarged negative bias of the global-mean SSS in IMPV1. It is argued that the slightly worse performance of IMPV1 of the global-mean SSS is attributed to the same unexpected negative bias in both IMPV0 and IMPV1 in the region adjacent to the original artificial island.

Fig. 2.
Fig. 2.

Time series of annual-mean, global area-weighted average of SST bias (°C) and SSS bias (psu) from Exps IMPV0 (blue curve) and IMPV1 (red curve) during years 1–650. The observations used for bias estimations are from the PHC3.0 dataset.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

Actually, IMPV1 improves both SST and SSS in the central Arctic Ocean (CAO)—that is, the region from 80° to 90°N—in terms of mean bias and root-mean-square error (RMSE). Relative to the observations from PHC 3.0, the area-weighted mean biases of SST (SSS) over the CAO in model years 601–650 are 0.76°C (−0.59 psu) in IMPV1, while they are 0.78°C (−0.61 psu) in IMPV0 (see the values for 5-m depth in Table 1). The SD of the area-weight mean SST (SSS) over the CAO are 0.01°C (0.01 psu). The RMSEs of SST (SSS) over the CAO in the same period are 0.80°C (0.85 psu) in IMPV1 and 0.81°C (0.89 psu) in IMPV0 (see the values for 5-m depth in Table 2). These numbers indicate that IMPV1 outperforms IMPV0 in the CAO in terms of both mean biases and RMSEs of SST and SSS. From the horizontal distributions of the errors of SST and SSS in both IMPV0 and IMPV1 (Figs. 3a,b and 3d,e) and their differences (Figs. 3c and 3f), significant improvements of both SST and SSS can be seen in the region between 30° and 120°E off the original artificial island in the CAO, which confirms the advantage of reducing the artificial island.

Table 1.

Area-weighted mean biases of temperature (°C) and salinity (psu) in the upper central Arctic Ocean, excluding the original artificial island in Exp IMPV0, relative to the observation from PHC3.0. The SD of area-weighted mean of temperature and salinity larger than 0.01°C and 0.01 psu are given in parentheses. The model output is taken from years 601–650; the values set in bold indicate the smaller bias of the two.

Table 1.
Table 2.

Area-weighted RMSEs of temperature (°C) and salinity (psu) in the upper central Arctic Ocean, excluding the original artificial island in Exp IMPV0, relative to the observation from PHC3.0. The model output is taken from years 601–650. Values set in bold indicate the smaller RMSE of the two.

Table 2.
Fig. 3.
Fig. 3.

(left) Annual-mean SST bias (°C) and (right) SSS bias (psu) in the Arctic Ocean of (top) Exps IMPV0 and (middle) IMPV1. The observations used for bias estimations are from the PHC3.0 dataset. (bottom) Differences between the two experiments, using finer contour intervals. The model output is taken from years 601–650.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

Furthermore, IMPV1 reduces or keeps the area-weighted mean biases and RMSEs of temperature and salinity in the upper CAO, with exceptions on the RMSEs of temperature at three depths: 35, 45, and 55 m (Tables 1 and 2).

b. Arctic Ocean circulation

As shown in Fig. 4 obtained from the AMAP (AMAP 1998), there are two major ice-water circulation systems in the upper Arctic Ocean—namely, the Beaufort Gyre, a clockwise gyre located north of Alaska; the Canadian Arctic Archipelago; the eastern part of Russia; and the transpolar drift, which carries the sea ice and water from the Laptev Sea and the East Siberian Sea across the North Pole to the Fram Strait. It should be noted that Fig. 4 is a schematic of the directions of the surface ocean currents, based on very sparse observations in the Arctic Ocean.

Fig. 4.
Fig. 4.

Annual-mean surface currents in the Arctic Ocean. This figure is a reproduction of Fig. 3.29 of AMAP (AMAP 1998).

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

The simulated ocean currents at the depth of 15 m in winter, summer, and its annual mean from Exps IMPV0 and IMPV1 are shown in Fig. 5. Each experiment captures the AMAP-observed spatial distribution of the Beaufort Gyre with a spatial scale a little larger than the observed. However, the locations of the center of the Beaufort Gyre differ in the two experiments. If we define the center of the Beaufort Gyre using the center of the innermost circle of clockwise vectors, then IMPV1 simulates a better location of the center than IMPV0 does, compared with the observed location that is about 20°–25°E of the date line (i.e., 155°–160°W), because the simulated center is located around the date line in IMPV0 (Figs. 5a, 5c, and 5e) and about 15°–20° east of the date line in IMPV1 (i.e., 160°–165°W; Figs. 5b, 5d, and 5f). In addition, the intensity distribution of ocean currents in the Beaufort Gyre is also an important feature. As suggested by observations, the upper-ocean current is very weak (about 0.01–0.05 m s−1). However, it is intensified to about 0.1 m s−1 immediately after it enters the region north of Alaska (Coachman and Barnes 1961; AMAP 1998). In each experiment, the intensification of the upper-ocean current north of Alaska is well simulated, which increases from about 0.01–0.03 to 0.07 m s−1.

Fig. 5.
Fig. 5.

Arctic Ocean currents in (left) winter, (middle) summer, and (right) its annual mean at 15-m depth from (top) Exps IMPV0 and (bottom) IMPV1. The arrows indicate the directions of currents, while the colors give their velocities (m s−1). The model output is taken from years 601–650.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

As for the simulation of the transpolar drift, both IMPV1 and IMPV0 reasonably reproduce its basic characteristics, which originate from the Laptev Sea and the East Siberian Sea, and finally exit through the Fram Strait, although the simulated tracks of the drift are a little different from the observed. Clearly, the reduced artificial island greatly decreases the distance between the drift and the North Pole, which is only 1° of latitude away from the North Pole in IMPV1 but 3° in IMPV0. To quantify the positive contribution of the reduced artificial island to the simulated transpolar drift, the annual-mean net volume fluxes (AMNVFs) across section W (a segment between 83° and 90°N along 85.5°W) and section E (a segment between 82° and 90°N along 94.5°E) by both IMPV0 and IMPV1 are compared (Fig. 6b), because the drift near the pole mainly consists of eastward ocean currents across section W and westward currents across section E (Fig. 6a). Qualitatively, both IMPV0 and IMPV1 produce a weaker eastward volume flux across section W and a westward volume flux across section E than the observation (cf. Figs. 5 and Figs. 6a). However, the reduction of the artificial island improves the simulation of these fluxes in IMPV1 quantitatively. The value of total AMNVFs across sections W and E in IMPV1 is 1.15 ± 0.07 Sv (1 Sv ≡ 106 m3 s−1), more than 1.03 ± 0.07 Sv in IMPV0 (Fig. 6b). To investigate the segment between 86° and 90°N across sections W and E, the value of AMNVFs across this segment is 0.30 ± 0.12 Sv in IMPV1, but it is −1.20 ± 0.18 Sv in IMPV0 (refer to the corresponding mean annual cycle in Fig. 6a). Obviously, the transport direction simulated by IMPV1 is consistent with the observation, while that by IMPV0 is wrong due to the impact of larger artificial island.

Fig. 6.
Fig. 6.

Mean annual cycles of the net volume fluxes in IMPV0 (blue curves) and IMPV1 (red curves) across (a) the region 86°–90°N across sections W and E, and (b) sections W and E. The units are in Sv. The positive sign is used for eastward volume flux across section W, and for the westward volume flux across section E. The model output is taken from years 601–650.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

We further calculate the net volume fluxes across some important straits/channels in/near the Arctic Ocean—that is, the Bering Sea Strait, Fram Strait, BSO, and BSX (Fig. 7). Figure 7a shows that IMPV1 and IMPV0 present very similar mean annual cycles of net volume fluxes across the Bering Strait, which reach the maxima of 1.74 and 1.78 Sv, respectively, in May. The simulated annual means by the two experiments are 1.21 ± 0.08 and 1.23 ± 0.09 Sv (northward), respectively. However, the observed annual mean is 0.80 ± 0.20 Sv over the period from 1990 to 2004 and the maximum transport of 1.30 Sv occurs in June (Woodgate et al. 2005). It means that both IMPV1 and IMPV0 overestimate the annual mean and maximum of the net volume flux across the Bering Strait. Especially, little contribution by the reduced artificial island is found for this strait. Actually, the northward transport in the Bering Strait is a combination of the Pacific–Arctic pressure head (driving the seawater northward) and the local wind effect (driving the seawater southward) according to Woodgate et al. (2005).

Fig. 7.
Fig. 7.

Mean annual cycles of net volume fluxes (Sv) in IMPV0 (blue curves) and IMPV1 (red curves) across the (a) Bering Strait, (b) Fram Strait, (c) BSO, and (d) BSX. The positive sign is used for the eastward/northward volume flux. The model output is taken from years 601–650.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

Different from the Bering Strait, the net volume flux across the Fram Strait is obviously influenced by the size of the artificial island (Fig. 7b). The annual-mean values of IMPV1 and IMPV0 are about −3.26 ± 0.09 and −3.41 ± 0.12 Sv, respectively, where the negative sign means the transport is southward. A recent literature suggested that the observed volume flux across the Fram Strait from 1980 to 2005 is about −1.7 Sv (interannual −4.7 to 0.3 Sv; Rudels et al. 2008; Beszczynska-Moeller et al. 2011). Clearly, both experiments overestimated the net volume flux across the Fram Strait, which can be attributed to the closed Nares Strait and the other straits/channels in the Canadian Arctic Archipelago in LICOM2 (which contributes about −1 Sv). This overestimation is partly reduced in IMPV1 due to the more realistic bathymetry in the Arctic Ocean.

The net volume flux across the BSO is also strongly influenced by the artificial island (Fig. 7c). With a more real bathymetry, IMPV1 simulates a weaker annual cycle, whose annual mean of 2.10 ± 0.11 Sv (eastward) is much closer to the observed value of 2.00 Sv (interannual 0.8–2.9 Sv) over the period from 1997 to 2007 (Smedsrud et al. 2010; Skagseth et al. 2008; Beszczynska-Moeller et al. 2011) than that by IMPV0 of 2.23 ± 0.07 Sv.

Different sizes of the artificial island also lead to different intensities of the net volume flux across the BSX. The annual mean in IMPV1 (1.55 ± 0.09 Sv, eastward) is much closer to the observed value of 2.00 ± 0.6 Sv (Gammelsrød et al. 2009) than that in IMPV0 (1.29 ± 0.08 Sv).

Through comparison of the Arctic Ocean circulation in these experiments, it is suggested that with a more realistic bathymetry IMPV1 better produces the Beaufort Gyre and transpolar drift. Besides, the net volume fluxes across the Fram Strait, BSO, and BSX are all improved in IMPV1 over IMPV0.

c. AMOC

The global meridional overturning circulation (GMOC) and AMOC streamfunctions simulated by IMPV0 and IMPV1 are shown in Fig. 8. Both experiments capture the basic patterns of the GMOC and AMOC streamfunctions with nearly the same strength in the region south of 20°N, but Exp IMPV1 with a more realistic bathymetry simulates stronger and deeper North Atlantic Deep Water (NADW) than Exp IMPV0. Compared with the estimated maximum of AMOC streamfunction in the North Atlantic Ocean that is about 18.0 ± 2.5 Sv (Lumpkin and Speer 2007), the maximum of 18.1 ± 0.99 Sv in IMPV1 outperforms that of 16.9 ± 0.93 Sv in IMPV0 (Fig. 8). Note that the latitudes of the maxima in both IMPV0 and IMPV1 are about 40°N, while that of the observation estimation is 24°N (Lumpkin and Speer 2007).

Fig. 8.
Fig. 8.

Annual-mean, zonally integrated overturning streamfunction (Sv) in (left) the global ocean and (right) the Atlantic Ocean from (top) Exp IMPV0 and (middle) Exp IMPV1. (bottom) Differences between the two experiments. The model output is taken from model years 601–650.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

In addition, because the NADW of IMPV1 goes deeper, the difference of meridional overturning circulation between the two experiments is very large in the Atlantic Ocean, which is about 2.5 Sv (Fig. 8f). Such a big difference could not be found in the rest of the global ocean (Fig. 8e). To explain why the AMOC strength of IMPV1 is stronger than the one of IMPV0, the barotropic streamfunction (BSF) and the mixed layer depth (MLD) for these two experiments are further investigated. The MLD is defined by a density difference criterion—that is, the density difference between the sea surface and MLD is 0.03 kg m−3. The subpolar gyre (SPG) can be divided into two parts—that is, the major part centered in the Labrador Sea and its northern extension centered in the Greenland–Iceland–Norwegian (GIN) Seas (Figs. 9a and 9b). The BSF of SPG in IMPV1 is 5.4 Sv stronger than that in IMPV0 near the Labrador Sea (Fig. 9c). As documented by Häkkinen and Rhines (2004) and DiNezio et al. (2009), the wind stress curl and the North Atlantic Oscillation (NAO) play the most important roles in modulating the strength of SPG. However, the difference of BSF between IMPV1 and IMPV0 could not be attributed to the wind or other forcing changes associated with NAO because the forcing for the two experiments is the same. The current from the Fram Strait plays a secondary role in maintaining the strength of the northern extension of SPG. Because of the slower current across the Fram Strait in IMPV1 (Fig. 7b), the vorticity of the northern extension of SPG in IMPV1 is smaller than the one in IMPV0. Therefore, most of the vorticity of SPG in IMPV1 locates more southern than the one in IMPV0, which leads to a stronger SPG centered in the Labrador Sea in IMPV1 (Fig. 9c). The consequence of the stronger SPG is an AMOC with more strength in IMPV1 (Böning et al. 2006; Gao and Yu 2008). In addition, the area-weighted MLD from 40° to 80°N in the North Atlantic Ocean of IMPV1 is 6 m deeper than that of IMPV0, which suggests that the deep water formation in IMPV1 is stronger (Fig. 9f).

Fig. 9.
Fig. 9.

(left) Annual-mean barotropic streamfunction (Sv) and (right) mixed layer depth (m) in the Atlantic Ocean from (top) Exp IMPV0 and (middle) Exp IMPV1. The differences between the two experiments are shown in the bottom. The model output is taken from model year 601–650.

Citation: Journal of Atmospheric and Oceanic Technology 31, 1; 10.1175/JTECH-D-13-00064.1

4. Summary and conclusions

Because of the numerical instability, the region composed of the North Pole and its two adjacent discrete zonal circles was treated as an artificial island in the standard version of LICOM2. In this study, efforts are made to reduce the size of the artificial island. The model stability is improved after proper vertical interpolation and analytic solving of the nonlinear equation in the turbulence scheme (Huang et al. 2014). The improved LICOM2 has a more realistic bathymetry in the Arctic Ocean—that is, only the North Pole is treated as the artificial island. The influence of the reduced artificial island is evaluated from 1) potential temperature and salinity in the Arctic Ocean, 2) the currents in the Arctic Ocean, 3) the net volume fluxes across some important straits in the Arctic Ocean, and 4) the AMOC streamfunction.

Two experiments with different bathymetry near the North Pole by LICOM2_imp show that the model with a more realistic bathymetry simulates better distributions of the temperature and salinity in the CAO. The two main circulation features of the Arctic Ocean—that is, the Beaufort Gyre (in terms of the location of its center) and the transpolar drift (in terms of its spatial pattern)—are also improved with the reduced artificial island. Besides, the biases of the net volume fluxes across the Fram Strait, BSO, and BSX are significantly reduced.

Finally, we have some suggestions for future improvements of LICOM2_imp. The artificial island at the North Pole should be removed completely. The Nares Strait and other straits/channels in the Canadian Arctic Archipelago should be opened, which will be beneficial for the simulation of net volume flux across the Fram Strait. The effect of sea ice formation, melting, and drifting should be explicitly included, which is helpful for the simulation of temperature distribution in the Arctic Ocean.

Acknowledgments

The discussion with Liu Hailong and Zhang Xuehong is very helpful to this work. This work is supported by the following funding: the National Basic Research Program of China (973 Program: Grants 2011CB309704, 2010CB951904, 2013CB956600), Joint Center for Global Change Studies (Grant 105019), and the National Natural Science Foundation of China (Grants 41023002, 41005053).

REFERENCES

  • AMAP, 1998: AMAP assessment report: Arctic pollution issues. Arctic Monitoring and Assessment Programme, 859 pp.

  • Antonov, J. I., , Locarnini R. A. , , Boyer T. P. , , Mishonov A. V. , , and Garcia H. E. , 2006: Salinity. Vol. 2, World Ocean Atlas 2005, NOAA Atlas NESDIS 62, 182 pp.

  • Arakawa, A., , and Lamb V. R. , 1977: Computational design of the basic dynamical processes of the UCLA general circulation model. Methods in Computational Physics, J. Chang, Ed., Vol. 17, Academic Press, 173–265.

  • Beszczynska-Moeller, A., , Woodgate R. A. , , Lee C. , , Melling H. , , and Karcher M. , 2011: A synthesis of exchanges through the main oceanic gateways to the Arctic Ocean. Oceanography, 24, 8299.

    • Search Google Scholar
    • Export Citation
  • Böning, C. W., , Scheinert M. , , Dengg J. , , Biastoch A. , , and Funk A. , 2006: Decadal variability of subpolar gyre transport and its reverberation in the North Atlantic overturning. Geophys. Res. Lett., 33, L21S01, doi:10.1029/2006GL026906.

    • Search Google Scholar
    • Export Citation
  • Bryan, K., , Manabe S. , , and Pacanowski R. C. , 1975: A global ocean–atmosphere climate model. Part II. The oceanic circulation. J. Phys. Oceanogr., 5, 3046.

    • Search Google Scholar
    • Export Citation
  • Canuto, V. M., , Howard A. , , Cheng Y. , , and Dubovikov M. S. , 2001: Ocean turbulence. Part I: One-point closure model—Momentum and heat vertical diffusivities. J. Phys. Oceanogr., 31, 14131426.

    • Search Google Scholar
    • Export Citation
  • Canuto, V. M., , Howard A. , , Cheng Y. , , and Dubovikov M. S. , 2002: Ocean turbulence. Part II: Vertical diffusivities of momentum, heat, salt, mass, and passive scalars. J. Phys. Oceanogr., 32, 240264.

    • Search Google Scholar
    • Export Citation
  • Canuto, V. M., , Howard A. , , Cheng Y. , , Muller, C. J. , Leboissetier A. , , and Jayne S. R. , 2010: Ocean turbulence, III: New GISS vertical mixing scheme. Ocean Modell., 34, 7091, doi:10.1016/j.ocemod.2010.04.006.

    • Search Google Scholar
    • Export Citation
  • Coachman, L. K., , and Barnes C. A. , 1961: The contribution of Bering Sea water to the Arctic Ocean. Arctic, 14, 147161.

  • Courant, R., , Friedrichs K. , , and Lewy H. , 1928: Über die partiellen Differenzengleichungen der mathematischen Physik. Math. Ann., 100, 3274.

    • Search Google Scholar
    • Export Citation
  • Coward, A. C., , Killworth P. D. , , and Blundell J. R. , 1994: Tests of a two-grid world ocean model. J. Geophys. Res., 99 (C11), 22 72522 735.

    • Search Google Scholar
    • Export Citation
  • Danabasoglu, G., , Bates S. C. , , Briegleb B. P. , , Jayne S. R. , , Jochum M. , , Large W. G. , , Peacock S. , , and Yeager S. G. , 2012: The CCSM4 ocean component. J. Climate, 25, 13611389.

    • Search Google Scholar
    • Export Citation
  • DiNezio, P. N., , Gramer L. J. , , Johns W. E. , , Meinen C. S. , , and Baringer M. O. , 2009: Observed interannual variability of the Florida Current: Wind forcing and the North Atlantic Oscillation. J. Phys. Oceanogr., 39, 721736.

    • Search Google Scholar
    • Export Citation
  • Eby, M., , and Holloway G. , 1994: Grid transformation for incorporating the Arctic in a global ocean model. Climate Dyn., 10, 241247.

  • Flato, G. M., , Boer G. J. , , Lee W. G. , , McFarlane N. A. , , Ramsden D. , , Reader M. C. , , and Weaver A. J. , 2000: The Canadian Centre for Climate Modelling and Analysis global coupled model and its climate. Climate Dyn., 16, 451467.

    • Search Google Scholar
    • Export Citation
  • Furevik, T., , Bentsen M. , , Drange H. , , Kindem I. K. T. , , Kvamstø N. G. , , and Sorteberg A. , 2003: Description and evaluation of the Bergen climate model: ARPEGE coupled with MICOM. Climate Dyn., 21, 2751.

    • Search Google Scholar
    • Export Citation
  • Gammelsrød, T., , Leikvin Ø. , , Lien V. , , Budgell W. P. , , Loeng H. , , and Maslowski W. , 2009: Mass and heat transports in the NE Barents Sea: Observations and models. J. Mar. Syst., 75, 5669.

    • Search Google Scholar
    • Export Citation
  • Gao, Y., , and Yu L. , 2008: Subpolar gyre index and the North Atlantic meridional overturning circulation in a coupled climate model. Atmos. Oceanic Sci. Lett., 1, 2932.

    • Search Google Scholar
    • Export Citation
  • Gent, P. R., , Bryan F. O. , , Danabasoglu G. , , Doney S. C. , , Holland W. R. , , Large W. G. , , and McWilliams J. C. , 1998: The NCAR Climate System Model global ocean component. J. Climate, 11, 12871306.

    • Search Google Scholar
    • Export Citation
  • Gordon, C., , Cooper C. , , Senior C. A. , , Banks H. , , Gregory J. M. , , Johns T. C. , , Mitchell J. F. B. , , and Wood R. A. , 2000: The simulation of SST, sea ice extents and ocean heat transports in a version of the Hadley Centre coupled model without flux adjustments. Climate Dyn., 16, 147168.

    • Search Google Scholar
    • Export Citation
  • Griffies, S. M., and Coauthors, 2000: Developments in ocean climate modelling. Ocean Modell., 2, 123192.

  • Griffies, S. M., and Coauthors, 2005: Formulation of an ocean model for global climate simulations. Ocean Sci., 1, 4579.

  • Häkkinen, S., , and Rhines P. B. , 2004: Decline of subpolar North Atlantic circulation during the 1990s. Science, 304, 555559.

  • Haney, R. L., 1971: Surface thermal boundary condition for ocean circulation models. J. Phys. Oceanogr., 1, 241248.

  • Huang, W.-Y., , Wang B. , , Yu Y.-Q. , , and Li L.-J. , 2014: Improvements in LICOM2. Part I: Vertical mixing. J. Atmos. Oceanic Technol., in press.

  • Kara, A. B., , Hurlburt H. E. , , and Wallcraft A. J. , 2005: Stability-dependent exchange coefficients for air–sea fluxes. J. Atmos. Oceanic Technol., 22, 10801094.

    • Search Google Scholar
    • Export Citation
  • Li, L. J., and Coauthors, 2013: The Flexible Global Ocean-Atmosphere-Land System Model, version g2: FGOALS-g2. Adv. Atmos. Sci., 543–560, doi:10.1007/s00376-012-2140-6.

    • Search Google Scholar
    • Export Citation
  • Liu, H.-L., , Lin P.-F. , , Yu Y.-Q. , , and Zhang X.-H. , 2012: The baseline evaluation of LASG/IAP climate system ocean model (LICOM) version 2. Acta Meteor. Sin., 26, 318329.

    • Search Google Scholar
    • Export Citation
  • Liu, X.-Y., , Liu H.-L. , , Zhang X.-H. , , and Yu R.-C. , 2006: Experiments of a reduced grid in LASG/IAP World Ocean general circulation models (OGCMs). J. Trop. Meteor., 12, 915.

    • Search Google Scholar
    • Export Citation
  • Locarnini, R. A., , Mishonov A. V. , , Antonov J. I. , , Boyer T. P. , , and Garcia H. E. , 2006: Temperature. Vol. 1, World Ocean Atlas 2005, NOAA Atlas NESDIS 61, 182 pp.

  • Lumpkin, R., , and Speer K. , 2007: Global ocean meridional overturning. J. Phys. Oceanogr., 37, 25502562.

  • Madec, G., 2008: NEMO ocean engine. IPSL Note du Pole de modélisation 27, 357 pp.

  • Mesinger, F., 1984: A blocking technique for representation of mountains in atmospheric models. Riv. Meteor. Aeronaut., 44, 195202.

  • Mesinger, F., , Janjić Z. I. , , Ničković S. , , Gavrilov D. , , and Deaven D. G. , 1988: The step-mountain coordinate: Model description and performance for cases of Alpine lee cyclogenesis and for a case of an Appalachian redevelopment. Mon. Wea. Rev., 116, 14931518.

    • Search Google Scholar
    • Export Citation
  • Murray, R. J., 1996: Explicit generation of orthogonal grids for ocean models. J. Comput. Phys., 126, 251273.

  • Murray, R. J., , and Reason C. J. C. , 2001: A curvilinear version of the Bryan–Cox–Semtner ocean model and its representation of the Arctic circulation. J. Comput. Phys., 171, 146.

    • Search Google Scholar
    • Export Citation
  • Murray, R. J., , and Reason C. J. C. , 2002: Fourier filtering and coefficient tapering at the North Pole in OGCMs. Ocean Modell., 4, 125.

    • Search Google Scholar
    • Export Citation
  • Pacanowski, R. C., , Dixon K. W. , , and Rosati A. , 1993: The GFDL Modular Ocean Model users guide. Geophysical Fluid Dynamics Laboratory Ocean Group Tech. Rep. 2, 46 pp.

  • Rudels, B., , Marnela M. , , and Eriksson P. , 2008: Constraints on estimating mass, heat and freshwater transports in the Arctic Ocean: An exercise. Arctic–Subarctic Ocean Fluxes, R. Dickson, J. Meincke, and P. Rhines, Eds., Springer Netherlands, 315–341.

  • Simmons, A. J., , and Gibson J. K. , 2000: The ERA-40 project plan. ERA-40 Project Report Series 1, ECMWF, 63 pp.

  • Skagseth, Ø., , Furevik T. , , Ingvaldsen R. , , Loeng H. , , Mork K. , , Orvik K. , , and Ozhigin V. , 2008: Volume and heat transports to the Arctic Ocean via the Norwegian and Barents Seas. Arctic–Subarctic Ocean Fluxes, R. Dickson, J. Meincke, and P. Rhines, Eds., Springer Netherlands, 45–64.

  • Smedsrud, L. H., , Ingvaldsen R. , , Nilsen J. E. Ø. , , and Skagseth Ø. , 2010: Heat in the Barents Sea: Transport, storage, and surface fluxes. Ocean Sci., 6, 219234.

    • Search Google Scholar
    • Export Citation
  • Smith, R., and Coauthors, 2010: The Parallel Ocean Program (POP) reference manual: Ocean component of the Community Climate System Model (CCSM). Los Alamos National Laboratory Tech. Rep. LAUR-10-01853, 141 pp.

  • Steele, M., , Morley R. , , and Ermold W. , 2001: PHC: A global ocean hydrography with a high-quality Arctic Ocean. J. Climate, 14, 20792087.

    • Search Google Scholar
    • Export Citation
  • Sun, H.-C., , and Zhou G.-Q. , 2010: An Arctic sea ice simulation using an ocean-ice coupled model. Atmos. Oceanic Sci. Lett., 3, 219223.

    • Search Google Scholar
    • Export Citation
  • Woodgate, R. A., , Aagaard K. , , and Weingartner T. J. , 2005: Monthly temperature, salinity, and transport variability of the Bering Strait through flow. Geophys. Res. Lett., 32, L04601, doi:10.1029/2004GL021880.

    • Search Google Scholar
    • Export Citation
  • Xu, S., and Coauthors, 2013: Simulation of sea ice in FGOALS-g2: Climatology and late 20th century changes. Adv. Atmos. Sci., 30, 658673.

    • Search Google Scholar
    • Export Citation
  • Yu, R., 1989: The design of a regional numerical forecasting model with steep topography. Sci. Atmos. Sin., 13, 139149.

  • Yu, R., 1994: A test for numerical weather prediction of real-time for China flood season precipitation in 1993 by a regional eta-coordinate model. Sci. Atmos. Sin., 18, 284292.

    • Search Google Scholar
    • Export Citation
  • Yu, Y.-Q., 1997: Design of ocean-sea ice-atmosphere coupling scheme and numerical simulation of interdecadal climate variability. Institute of Atmospheric Physics, Chinese Academy of Sciences, 130 pp.

Save