Accurate representation of the Atlantic Meridional Overturning Circulation (AMOC) in global climate models is crucial for reliable future climate predictions and projections. In this study, we used 42 coupled atmosphere–ocean global climate models to analyze low-frequency variability of the AMOC driven by the North Atlantic Oscillation (NAO). Our results showed that the influence of the simulated NAO on the AMOC differs significantly between the models. We showed that the large intermodel diversity originates from the diverse oceanic mean state, especially over the subpolar North Atlantic (SPNA), where deep water formation of the AMOC occurs. For some models, the climatological sea ice extent covers a wide area of the SPNA and restrains efficient air–sea interactions, making the AMOC less sensitive to the NAO. In the models without the sea-ice-covered SPNA, the upper-ocean mean stratification critically affects the relationship between the NAO and AMOC by regulating the AMOC sensitivity to surface buoyancy forcing. Our results pinpoint the oceanic mean state as an aspect of climate model simulations that must be improved for an accurate understanding of the AMOC.

Using large-eddy simulations (LES) it is shown that the depth of a diurnal thermocline h should be scaled by the Zilitinkevich scale LZ, not by the Monin–Obukhov length scale LMO, contrary to the proposition by Pearson et al. Their argument to explain the slower increase of h than LMO using the effect of the preexisting thermocline is also invalid.

The boundary layers of the atmosphere and the ocean are compared during convection, and their latitudinal dependence is investigated. The results are applied to examine the parameterization of the boundary layer depth in the K‐profile parameterization (KPP) model. The bulk Richardson number Rib varies excessively with time in the high‐latitude (HL) oceanic boundary layer (OBL) without unresolved shear Vt2, as a result of inertial oscillation. Inclusion of Vt2 is also necessary in the atmospheric boundary layer (ABL) to mitigate the large variation of Rib with wind stress. Stratification and velocity shear near the boundary layer height/depth are stronger and thicker at low latitudes (LLs), where the Ekman length scale is larger and the inertial time scale is longer. This enhanced shear makes the entrainment buoyancy flux larger at LL. Analysis of the turbulent kinetic energy (TKE) budget in the entrainment zone is carried out to explain the variation of Rib depending on the boundary layer and the latitude. This analysis shows that the TKE production in the entrainment zone is dominated by shear production at LL, but by the TKE flux at HL, and that Vt2 represents the contribution from the TKE flux to the entrainment zone. The enhancement of vertical TKE by Langmuir circulation (LC) does not depend on the latitude, but it decreases with depth faster at LL. The result suggests that the parameterization of the boundary layer depth in the KPP model should be different depending on whether it is the ABL or the OBL and depending on the latitude.

No abstract

The ocean mixed layer model (OMLM) is improved using the large eddy simulation (LES) and the inverse estimation method. A comparison of OMLM (Noh model) and LES results reveals that underestimation of the turbulent kinetic energy (TKE) flux in the OMLM causes a negative bias of the mixed layer depth (MLD) during convection, when the wind stress is weak or the latitude is high. It is further found that the entrainment layer thickness is underestimated. The effects of alternative approaches of parameterizations in the OMLM, such as nonlocal mixing, length scales, Prandtl number, and TKE flux, are examined with an aim to reduce the bias. Simultaneous optimizations of empirical constants in the various versions of Noh model with different parameterization options are then carried out via an iterative Green’s function approach with LES data as constraining data. An improved OMLM is obtained, which reflects various new features, including the enhanced TKE flux, and the new model is found to improve the performance in all cases, namely, wind-mixing, surface heating, and surface cooling cases. The effect of the OMLM grid resolution on the optimal empirical constants is also investigated.