Abstract

This study shows that a reduction in vertical mixing applied inside the Atlantic basin can drastically increase North Atlantic Deep Water (NADW) stability with respect to freshwater perturbations applied to the North Atlantic. This is contrary to the notion that the stability of the ocean’s thermohaline circulation simply scales with vertical mixing rates. An Antarctic Intermediate Water (AAIW) reverse cell, reliant upon upwelling of cold AAIW into the Atlantic thermocline, is found to be associated with stable states where NADW is collapsed. Transitions between NADW “on” and “off” states are characterized by interhemispheric competition between this AAIW cell and the NADW cell. In contrast to the AAIW reverse cell, NADW eventually upwells outside the Atlantic basin and is thus not as sensitive to changes in vertical mixing within the Atlantic. A reduction in vertical mixing in the Atlantic weakens the AAIW reverse cell, resulting in an enhanced stability of NADW formation. The results also suggest that the AAIW reverse cell is responsible for the stability of NADW collapsed states, and thereby plays a key role in maintaining multiple equilibria in the climate system. A global increase of vertical mixing in the model results in significantly enhanced NADW stability, as found in previous studies. However, an enhancement of vertical mixing applied only inside the Atlantic Ocean results in a reduction of NADW stability. It is concluded that the stability of NADW formation to freshwater perturbations depends critically on the basin-scale distribution of vertical mixing in the world’s oceans.

1. Introduction

Stommel (1961) first proposed the possibility that the ocean’s thermohaline circulation could have two stable regimes of flow. Bryan (1986) demonstrated how the operation of a positive salt feedback could bring about two asymmetric overturning states under symmetric surface flux conditions in a rectangular basin geometry. In his idealized model, there is competition for interhemispheric freshwater (FW) export between two overturning cells involving the sinking of water originating at the antipodean polar surface of the basin. The real ocean, however, contains an obstruction to southward geostrophic flow across the latitudes of Drake Passage (DP), thus inhibiting the formation of the vigorous Southern Hemisphere (SH) cell observed by Bryan (1986). Instead, studies employing a more realistic geometry (e.g., Manabe and Stouffer 1988, 1999, hereafter MS99; Saenko et al. 2003; Gregory et al. 2003; Rahmstorf 1996; Sijp and England 2005) find that SH overturning states exhibit a shallower Antarctic Intermediate Water (AAIW) reverse cell inside the Atlantic basin. MS99 point out that this small reverse cell is associated with a stable North Atlantic Deep Water (NADW) “off” state. In contrast to the SH cell of Bryan (1986) and other studies employing a Drake Passage closed geometry (e.g. Mikolajewicz et al. 1993; Sijp and England 2004), this AAIW reverse cell originates at the AAIW formation regions. Saenko et al. (2003) use a model to suggest that due to the existence of this cell in an NADW off state, the density difference between surface waters at the formation regions for AAIW and NADW determines the strength and polarity of the meridional overturning circulation (MOC), at least in their model. Sijp and England (2005) show that the nature of this relationship depends on the depth of the Drake Passage sill, with Antarctic Bottom Water (AABW) playing a reduced role as the DP deepens. Furthermore, Saenko et al. (2003) and Gregory et al. (2003) suggest that this AAIW reverse cell can contribute to maintaining a stable NADW off state by importing FW into the Atlantic basin, and that it cannot coexist with the NADW formation cell.

In ocean models the upwelling branches of the NADW cell and the AAIW reverse cell rely in part on vertical mixing rates at lower latitudes. This upwelling into the thermocline consists of a balance between downward diffusion of heat from the surface and upward advection of colder water from below the thermocline (e.g., Munk 1966; Munk and Wunsch 1998), the balance determining the global thermocline depth. Enhanced vertical mixing therefore increases transport in the upwelling branches of these cells. Indeed, Bryan (1987) found that larger vertical mixing rates lead to increased MOC in an idealized sector geometry model. There have been several other studies that investigate the role of vertical mixing in setting the thermohaline circulation (THC) strength in global ocean climate models. MS99 showed that the stability of their NADW off state depends on the value of vertical mixing. They also find enhanced NADW formation and greater NADW stability with respect to FW perturbations under a global increase in vertical mixing. Studies employing a zonally averaged model (Schmittner and Weaver 2001) and a three-dimensional ocean model (Prange et al. 2003) use hysteresis experiments to show that the stability of the NADW off and “on” states depend on the rate of vertical mixing. For instance, in agreement with MS99, Schmittner and Weaver (2001) find enhanced NADW stability in their hysteresis experiments when vertical mixing is increased globally. These studies, however, employ horizontally uniform vertical mixing modifications. In the present study, we will assess the stability of the North Atlantic MOC with respect to basinwide changes in the vertical mixing coefficient (Kυ).

The AAIW reverse cell of the NADW off state relies on upwelling inside the Atlantic basin, whereas in the NADW on state, NADW outflow leaves the Atlantic basin at 30°S to upwell predominantly outside this basin. Broecker (1991) highlights the role of the Indian and Pacific Oceans in the removal of NADW by vertical mixing at low latitudes in the classical “global ocean conveyor belt” schematic. Later studies (e.g., Toggweiler and Samuels 1995; Döös 1995) suggest instead that a significant portion of NADW resurfaces via wind-driven upwelling in the Southern Ocean, undergoing subsequent buoyancy changes due to surface fluxes. Model studies employing weak background vertical mixing also support this idea (e.g., Saenko and Merryfield 2005). Another example of how a reasonable overturning and a thermocline of realistic depth can be achieved without the need for strong vertical diffusivity is described in the theory of the ventilated thermocline (Luyten et al. 1983).

As the upwelling branches of the NADW cell and the AAIW reverse cell occur at different locations, the spatial distribution of vertical mixing in the World Ocean will likely affect the relative potential of these cells for dominance of deep ocean ventilation. In particular, the Atlantic AAIW reverse cell relies upon the removal of intermediate water across the Atlantic thermocline via diapycnal mixing. Hence, a reduction in vertical mixing at low latitudes of the Atlantic should reduce the strength of the AAIW reverse cell and therefore its ability to compete with NADW. The NADW cell, on the other hand, does not rely on Atlantic upwelling and may therefore be less sensitive to a reduction of vertical mixing inside the Atlantic. In this situation, a greater robustness of NADW overturning with respect to FW perturbations may ensue. In this study, we use an intermediate-complexity coupled climate model to examine the stability of NADW formation with respect to FW perturbations under different spatial distributions of vertical mixing.

The remainder of this paper is divided as follows. Section 2 covers a description of the model and experimental design. We will consider experiments where Kυ is changed only inside the Atlantic Ocean to examine the role of the AAIW reverse cell in suppressing NADW formation. We will also assess experiments wherein a reduced Kυ is applied only inside the Indian and Pacific Oceans. To compare our findings to previous studies, we further present an experiment similar to MS99 where Kυ is increased globally, and an experiment where Kυ is increased by a similar magnitude, but only inside the Atlantic. In section 3 we first discuss the steady state fields under the different vertical mixing scenarios, and then describe their response to the application of external FW pulses of varying magnitude. Finally, section 4 covers a discussion and the conclusions.

2. Model and numerical experiments

We use the University of Victoria Intermediate Complexity Coupled Model described in detail in Weaver et al. (2001), which comprises a global ocean general circulation model [Geophysical Fluid Dynamics Laboratory (GFDL) Modular Ocean Model (MOM) Version 2.2; Pacanowski 1995] coupled to a simplified atmospheric model and a dynamic–thermodynamic sea ice model. A global domain is used with horizontal resolution of 1.8° latitude by 3.6° longitude in each model component. While air–sea heat and freshwater fluxes evolve freely in the model, a noninteractive wind field is employed. FW fluxes between the ocean and the atmosphere are determined by evaporation and precipitation, while river runoff and changes in sea ice volume also affect oceanic salinity. Moisture transport in the atmosphere occurs by way of advection and diffusion. Precipitation occurs when relative humidity exceeds a threshold value. The wind forcing is taken from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis fields (Kalnay et al. 1996), averaged over the period 1958–97 to form a seasonal cycle from the monthly fields. Brine rejection during sea ice formation is parameterized after Duffy and Caldeira (1997). The atmospheric model advects and diffuses moisture, and it diffuses heat. No flux corrections are used (for further details, see Weaver et al. 2001). Vertical mixing in the control experiment is achieved using a horizontally uniform diffusivity that increases with depth, taking a value of 0.6 cm2 s−1 at the surface and increasing to 1.6 cm2 s−1 at the bottom. The effect of subgrid-scale eddies on horizontal tracer transport is modeled by globally uniform horizontal diffusion. The diffusion coefficient in both horizontal directions is Ah = 2 × 107 cm2 s−1. The constant lateral mixing coefficient for momentum is 2 × 109 cm2 s−1, and the constant vertical mixing coefficient for momentum is 10 cm2 s−1. For economy of computation, with many multimillennial integrations, no parameterization of along-isopycnal diffusion (Redi 1982) or eddy-induced advective tracer transport (Gent and McWilliams 1990, hereafter GM) is used in the standard set of experiments. Rather, our focus is on model THC sensitivity to vertical diffusive mixing. It is noted, however, that we have reevaluated several of the experiments under GM and confirm that our results are robust in both GM and non-GM experiments.

To examine the effect of reducing Kυ in different ocean basins, we have integrated the following configurations of the model to equilibrium: 1) a control experiment CNTRL, where no reduction in Kυ is applied; 2) an experiment denoted ⅔KVAtl, where Kυ takes a value reduced to two-thirds of CNTRL inside the Atlantic basin; 3) an experiment denoted ⅓KVAtl, where Kυ is multiplied inside the Atlantic basin by a factor of 1/3; and 4) an experiment where Kυ is similarly reduced inside the Pacific and Indian basins, denoted ⅓KVIP. Figure 1 shows the vertical profile of Kυ in these experiments. Figure 2a shows the two areas where reduction of Kυ is applied in the respective experiments. The reduction is only applied between 35°S and 48°N in each ocean basin. The above reductions in Kυ are also only applied between the surface and 1257-m depth, with no change in Kυ at levels deeper than 1257 m (Fig. 1). A version of ⅓KVAtl using GM has been run to verify the robustness of our results with respect to the choice of parameterization of subgrid-scale eddy mixing of tracer properties.

Fig. 1.

Vertical mixing profile of CNTRL, ⅔KVAtl, (1/3)KVAtl [equivalent to that used in ⅓KVIP (solid) only over a different subdomain], and the MS99KV experiments.

Fig. 1.

Vertical mixing profile of CNTRL, ⅔KVAtl, (1/3)KVAtl [equivalent to that used in ⅓KVIP (solid) only over a different subdomain], and the MS99KV experiments.

Fig. 2.

(a) Model domain and areas of the Atlantic (Atl, dark gray) and Indian/Pacific Oceans (IP, light gray) where Kυ is reduced. (b) FW perturbation vs time. The region of FW addition/extraction is indicated in (a) by the region enclosed by solid lines. A maximum value M is attained after 150 yr. The duration of all perturbations is 300 yr. No values are indicated along the vertical axis as perturbations of different magnitudes are used, ranging from −0.48 to 0.76 Sv.

Fig. 2.

(a) Model domain and areas of the Atlantic (Atl, dark gray) and Indian/Pacific Oceans (IP, light gray) where Kυ is reduced. (b) FW perturbation vs time. The region of FW addition/extraction is indicated in (a) by the region enclosed by solid lines. A maximum value M is attained after 150 yr. The duration of all perturbations is 300 yr. No values are indicated along the vertical axis as perturbations of different magnitudes are used, ranging from −0.48 to 0.76 Sv.

To further examine whether our model exhibits an increase in NADW stability upon increasing Kυ globally at a depth range containing the pycnocline, as found by MS99, we have integrated an experiment MS99KVglob similar to their so-called large vertical diffusion (LVD) experiment, where Kυ takes an increased global value of 1.3 cm2 s−1 down to 2500-m depth (see Fig. 1). Below 2500-m depth Kυ follows the profile of CNTRL. The MS99 LVD experiment employs constant Kυ = 1.3 cm2 s−1; however, such a profile would yield a decrease in Kυ at depth compared to CNTRL. Hence, the profile shown in Fig. 1 is adopted for direct comparison with the CNTRL experiment. To elucidate the role of vertical mixing inside the Atlantic in the context of the experiments of MS99, we have also integrated an experiment MS99KVAtl, where Kυ is changed in a similar fashion to MS99KVglob, but only inside the Atlantic. The above experiments should yield similar results to experiments using the actual MS99 profile with Kυ = 1.3 cm2 s−1 throughout the water column, as diapycnal mixing inside the pycnocline, rather than abyssal mixing, is the main factor affecting the strength of NADW formation. To confirm this, we have run experiments using a value of 1.3 cm2 s−1 Kυ throughout the water column, as in MS99, and found results that are very similar to those of MS99KVglob and MS99KVAtl. Hence, only the one set of experiments with increased Kυ are reported here.

We further subject the obtained equilibria for CNTRL, ⅔KVAtl, ⅓KVAtl, ⅓KVIP, MS99KVglob and MS99KVAtl to FW pulses applied to the North Atlantic (NA) for the duration of 300 yr. We vary the maximum magnitude of this pulse. In short, we vary both Kυ and the magnitude of the FW pulse in this study, and all other model parameters remain fixed. The FW perturbations are applied in the NA between the black lines shown in Fig. 2a. Figure 2b shows the FW perturbation against time. A linear increase from 0 to a maximum value M occurs over the first 150 yr, followed by a linear decline back to 0 over the following 150 yr. After year 300, no further FW perturbation is applied. Note that we apply this perturbation for several values of M. This procedure, similar to that of Sijp and England (2005), is designed to examine the existence of multiple equilibria in the respective experiments.

The ocean exhibits substantial regional variations in Kυ (e.g., Ledwell et al. 2000), thought to be a function of tidal currents, subsurface bathymetry, and local stratification. Several studies have examined the effects of locally enhanced vertical diffusivity over rough bathymetry in ocean models (e.g., Hasumi and Suginohara 1999; Saenko and Merryfield 2005). Unlike these previous studies, we make no attempt to examine the effects of contrasts in vertical mixing that may exist between basins in the real ocean. Instead our experiments are specifically aimed to examine the role of the location of deepwater removal in determining the global THC and its stability.

3. Results

a. Equilibrated states

In this section we first analyze the steady states of the experiments. After discussing these equilibria we will present a series of perturbations to these steady states where the maximum value of the perturbation varies (section 3b). Figure 3 shows the MOC in the Atlantic basin for the NADW on states and NADW off states of experiments CNTRL and ⅔KVAtl. Also shown are the steady NADW on state of ⅓KVAtl and the only steady state found for ⅓KVIP. A summary of the MOC transport rates for all experiments is given in Table 1. The NADW on state in CNTRL exhibits 21.5 Sv (1 Sv ≡ 106 m3 s−1; throughout this paper all transport values are quoted to the nearest 0.1 Sv) of NADW formation. The CNTRL NADW off state was obtained from the NADW on state by applying a perturbation of maximum value M = 0.11 Sv (0.38 m yr−1). This NADW off state exhibits an AAIW reverse cell of 9.3 Sv. This cell overlies an AABW cell recirculating below 2000-m depth with an AABW inflow into the Atlantic sector of 4.5 Sv.

Fig. 3.

Atlantic meridional overturning streamfunction (annual mean) for (a) the steady NADW on state of the control experiment CNTRL, (b) the steady NADW off state of CNTRL, (c) the steady NADW on state of ⅔KVAtl, (d) the steady NADW off state of ⅔KVAtl, (e) the steady NADW on state of ⅓KVAtl, and (f) the steady state found in ⅓KVIP. Values are given in Sv. The states shown in (e) and (f) were the only stable equilibria found for those experiments in this study.

Fig. 3.

Atlantic meridional overturning streamfunction (annual mean) for (a) the steady NADW on state of the control experiment CNTRL, (b) the steady NADW off state of CNTRL, (c) the steady NADW on state of ⅔KVAtl, (d) the steady NADW off state of ⅔KVAtl, (e) the steady NADW on state of ⅓KVAtl, and (f) the steady state found in ⅓KVIP. Values are given in Sv. The states shown in (e) and (f) were the only stable equilibria found for those experiments in this study.

Table 1.

Summary of the main features of the meridional mass transport streamfunction in Sv for all model experiments. Shown are the NADW formation and outflow rate for the NADW on states, and the AAIW reverse cell for the NADW off states. We have been unable to find a stable NADW off state for ⅓KVAtl, and therefore no AAIW reverse cell strength is indicated. Similarly, no values associated with the NADW on state are shown for ⅓KVIP. The final column shows whether we find an increased or reduced stability of NADW formation with respect to FW perturbations in each experiment relative to CNTRL.

Summary of the main features of the meridional mass transport streamfunction in Sv for all model experiments. Shown are the NADW formation and outflow rate for the NADW on states, and the AAIW reverse cell for the NADW off states. We have been unable to find a stable NADW off state for ⅓KVAtl, and therefore no AAIW reverse cell strength is indicated. Similarly, no values associated with the NADW on state are shown for ⅓KVIP. The final column shows whether we find an increased or reduced stability of NADW formation with respect to FW perturbations in each experiment relative to CNTRL.
Summary of the main features of the meridional mass transport streamfunction in Sv for all model experiments. Shown are the NADW formation and outflow rate for the NADW on states, and the AAIW reverse cell for the NADW off states. We have been unable to find a stable NADW off state for ⅓KVAtl, and therefore no AAIW reverse cell strength is indicated. Similarly, no values associated with the NADW on state are shown for ⅓KVIP. The final column shows whether we find an increased or reduced stability of NADW formation with respect to FW perturbations in each experiment relative to CNTRL.

The experiment ⅔KVAtl admits an NADW on state, as shown in Fig. 3c, although with a reduction of NADW recirculation within the Atlantic and an increase in NADW outflow. We obtain the NADW off state in ⅔KVAtl from the NADW on state in a similar fashion to the CNTRL experiment. In the NADW off state of experiment ⅔KVAtl the AAIW reverse cell is reduced by 3 Sv, taking a value of 6.3 Sv, and is restricted to shallower depths (Fig. 3d). This reduction in the strength and depth of the AAIW reverse cell results from a decrease in AAIW upwelling into the Atlantic thermocline due to lower vertical mixing rates there. This result illustrates the dependence of the AAIW reverse cell on vertical mixing inside the Atlantic basin. In contrast, the AABW recirculation in ⅔KVAtl is similar (4.3 Sv) to that of the NADW off state in CNTRL (4.5 Sv). This could be because AABW inflow into the Atlantic returns south at depth and is not affected by changes in vertical mixing applied in the upper 1257 m, and changes in NADW formation are only small. The NADW on state in ⅓KVAtl exhibits a reduction of NADW formation from 21.5 Sv in CNTRL to 19.8 Sv. It is interesting to note that despite a reduction in NADW formation, NADW outflow1 is increased from 12.2 Sv in CNTRL to 14.8 Sv in ⅓KVAtl. This implies a reduction of NADW recirculation inside the Atlantic2 from 9.3 Sv in the control experiment to 5.0 Sv when vertical mixing is reduced in ⅓KVAtl. Thus, as expected, the rate of NADW recirculation inside the Atlantic basin depends on the rate of localized vertical mixing. A reduction of vertical mixing in the Atlantic therefore reduces this recirculation, which is compensated in part by reduced NADW formation, and in part by an increase in NADW outflow.

NADW formation is absent in the steady-state MOC in ⅓KVIP, despite being set up identical to CNTRL apart from a reduction of Kυ in the tropical Indian and Pacific Oceans. This suggests that the maintenance of an NADW overturning circulation depends in part on vertical mixing rates in the ocean basins outside the Atlantic. For ⅓KVIP, the Atlantic MOC is similar to that of the NADW off state of CNTRL (cf. Figs. 3b,f), with an AAIW reverse cell of 10.2 Sv overlying an AABW cell recirculating below 2000-m depth. This suggests that the Atlantic circulation of the NADW off state in CNTRL is not significantly influenced by vertical mixing outside the Atlantic basin. The similarity between the CNTRL NADW off state and the unperturbed ⅓KVIP, experiment arises because the AAIW reverse cell relies almost exclusively on vertical mixing inside the Atlantic for its upwelling branch.

In contrast to experiments CNTRL and ⅔KVAtl, we were unable to obtain stable NADW off states in ⅓KVAtl. This was attempted with additions of FW perturbations of up to M = 0.58 Sv. This shows a significant increase in NADW robustness with respect to FW perturbations in the NA when Kυ is reduced in the Atlantic sector. This is because a strong reduction of Kυ in the Atlantic weakens the AAIW reverse cell that is otherwise required to prevent the eventual reestablishment of the NADW cell after the forcing is removed. As we will see later, the reverse cell appears temporarily during our FW perturbation, but with reduced vigor in ⅓KVAtl. Unlike ⅓KVAtl, the existence of a stable NADW off state in the ⅔KVAtl experiment allows us to examine the effect of vertical mixing in the Atlantic on the AAIW reverse cell.

To test the relationship between the AAIW reverse cell and the value of Kυ inside the Atlantic, we have run several experiments similar to ⅓KVAtl, only using multiplicative factors of 0.75, 0.85, and 1.25. In all cases we obtain NADW off states using an FW perturbation. Figure 4 shows the strength of the AAIW reverse cell in the NADW off states against the multiplicative factor applied to Kυ for these experiments, along with ⅔KVAtl and CNTRL. The solid line is a linear “best fit” using the method of least squares. The close fit of this line to our experimental results suggests that the strength of the AAIW reverse cell has a linear dependence on the rate of vertical mixing inside the Atlantic, as would arise if a simple advection–diffusion balance were in place. From this analysis we can assume that vertical mixing inside the Atlantic constitutes a limiting factor on the strength of the AAIW reverse cell.

Fig. 4.

AAIW reverse cell in NADW off states vs the multiplicative factor applied to Kυ in the Atlantic Ocean. As shown in Fig. 1, the multiplicative factor is only applied to Kυ in the upper 1257-m depth. The solid line is a linear best fit using the method of least squares. Values are given in Sv.

Fig. 4.

AAIW reverse cell in NADW off states vs the multiplicative factor applied to Kυ in the Atlantic Ocean. As shown in Fig. 1, the multiplicative factor is only applied to Kυ in the upper 1257-m depth. The solid line is a linear best fit using the method of least squares. Values are given in Sv.

MS99 found an increased stability of NADW when globally increasing Kυ to 1.3 cm2 s−1 in their so-called LVD experiment. This result appears to be in contradiction to our finding that decreasing Kυ inside the Atlantic results in an increase in NADW stability. Therefore, we now examine the results of our experiment MS99KVglob, where we set Kυ to 1.3 cm2 s−1 in the upper 2500 m globally, similar in design to the LVD experiment of MS99. A second experiment MS99KVAtl is evaluated where we apply this increase of Kυ only inside the Atlantic. Note that our CNTRL experiment employs a value of 0.6 cm2 s−1 at the top layer. Experiments MS99KVglob and MS99KVAtl therefore employ larger values within the global and Atlantic pycnocline, respectively, than our CNTRL experiment. Figures 5a–c show the MOC in the Atlantic basin for the NADW on state of MS99KVAtl, the NADW off state of MS99KVAtl, and the NADW on state of MS99KVglob, respectively. Increasing Kυ only inside the Atlantic (MS99KVAtl) results in a 2.2-Sv increase in NADW formation to 23.7 Sv and yet a significant reduction in NADW outflow (from 12.2 Sv down to 7.8 Sv) for the NADW on state (Table 1). Enhancing vertical mixing inside the Atlantic therefore results in a significant increase in NADW recirculation within the Atlantic at the expense of NADW outflow. AABW inflow remains at similar values to CNTRL. As expected, the NADW off state for MS99KVAtl (Fig. 5b) shows a significantly increased AAIW reverse cell, attaining a strength of 15.1 Sv. In agreement with the results shown in Fig. 4, this indicates that the strength of the AAIW reverse cell is set by vertical mixing inside the Atlantic. Enhancing Kυ only inside the Atlantic results in a significant increase in strength of the AAIW reverse cell due to the advective–diffusive balance that the upwelling branch of this cell relies on. Strong overturning (28.0 Sv) occurs in MS99KVglob (Fig. 5c), and the streamfunction in this experiment is very similar to that shown in Fig. 9a of MS99. Finally, it is noted that in agreement with MS99, we have been unable to obtain a stable NADW off state for MS99KVglob.

Fig. 5.

Atlantic meridional overturning streamfunction (annual mean) for (a) the steady NADW on state of MS99KVAtl, (b) the steady NADW off state of MS99KVAtl, and (c) the steady NADW on state of MS99KVglob. Values are given in Sv. The overturning shown in (c) is the only stable equilibrium found in the MS99KVglob experiment.

Fig. 5.

Atlantic meridional overturning streamfunction (annual mean) for (a) the steady NADW on state of MS99KVAtl, (b) the steady NADW off state of MS99KVAtl, and (c) the steady NADW on state of MS99KVglob. Values are given in Sv. The overturning shown in (c) is the only stable equilibrium found in the MS99KVglob experiment.

b. Freshwater perturbations

As stated in section 2 we have subjected the equilibria obtained for all experiments to 300-yr FW pulses of different magnitude. We also apply FW pulses to experiments with Kυ multiplied by factors of 1/4 and 1/2 in the Atlantic (denoted ¼KVAtl and ½KVAtl, respectively). Figures 6 –9 summarize the results. Figure 6 shows the time series of NADW formation and the AAIW reverse cell for experiments ¼KVAtl, ⅓KVAtl, ½KVAtl, and ⅔KVAtl under an FW perturbation attaining a maximum value M of 0.29 Sv (1.02 m yr−1). In other words, in this set of experiments, the maximum value M of the FW pulse is fixed, and Kυ varies. The FW pulse is sufficient to shut down NADW formation in ⅔KVAtl, whereas in ⅓KVAtl and ¼KVAtl a recovery of NADW formation occurs after the pulse has ceased. In ½KVAtl we see a slower recovery of NADW formation and an associated decline of the AAIW reverse cell after the FW pulse has ceased. This set of experiments shows that decreasing Kυ in the Atlantic results in an increased robustness of NADW formation with respect to a given FW pulse.

Fig. 6.

Time series of (a) NADW production rate and (b) the AAIW reverse cell for experiments ¼KVAtl (solid), ⅓KVAtl (dashed), ½KVAtl (dashed–dotted), and ⅔KVAtl (dotted) under an FW perturbation attaining a maximum value of 0.29 Sv (1.02 m yr−1) after 150 yr. Values are given in Sv.

Fig. 6.

Time series of (a) NADW production rate and (b) the AAIW reverse cell for experiments ¼KVAtl (solid), ⅓KVAtl (dashed), ½KVAtl (dashed–dotted), and ⅔KVAtl (dotted) under an FW perturbation attaining a maximum value of 0.29 Sv (1.02 m yr−1) after 150 yr. Values are given in Sv.

Fig. 9.

Time series of (a) NADW production rate and (b) the AAIW reverse cell for experiment CNTRL under a perturbation attaining a maximum value of 0.1 Sv (0.34 m yr−1). Values are given in Sv.

Fig. 9.

Time series of (a) NADW production rate and (b) the AAIW reverse cell for experiment CNTRL under a perturbation attaining a maximum value of 0.1 Sv (0.34 m yr−1). Values are given in Sv.

Figure 7 shows time series of NADW formation and the AAIW reverse cell in ⅓KVAtl in response to the application of a series of 300-yr FW pulses of varying magnitude M. Thus, in this analysis, we have kept the reduction of Kυ fixed while varying the magnitude M of the FW flux. For comparison, we have included the response of CNTRL to a smaller FW pulse. A perturbation of maximum value M = 0.11 Sv (0.38 m yr−1) yields a stable NADW off state in the control experiment. In contrast, when FW pulses attaining higher maximum values of M up to 0.43 Sv (1.53 m yr−1) are applied to ⅓KVAtl, NADW formation eventually recovers (Fig. 7a). The NADW shutdown in CNTRL coincides with the emergence of the AAIW reverse cell (Fig. 7b), which eventually attains a value of 9.3 Sv. When Kυ is reduced in the Atlantic, however, the role of the AAIW reverse cell appears to be diminished. Despite the initial suppression of the NADW cell in response to the FW perturbations applied to ⅓KVAtl, and the temporary establishment of a higher maximum strength AAIW reverse cell in one of the experiments, there is no permanent sustenance of this cell. After the perturbation has terminated, NADW formation recovers over a period of 1500–2500 yr. This indicates that in the absence of an external FW flux, circulation changes alone cannot maintain the light surface conditions at the NADW formation regions required to prevent sinking when Kυ is reduced by a factor of 1/3 in the Atlantic. Indeed, Gregory et al. (2003) suggest that the AAIW reverse cell is responsible for importing FW into the Atlantic, maintaining buoyant NA surface conditions that prevent a transition to an NADW on state. In contrast to the AAIW reverse cell, NADW finds its upwelling areas outside the Atlantic basin and is thus not subject to the same impediment as the AAIW reverse cell when Kυ is reduced in ⅓KVAtl. This indicates that a reduction of Kυ inside the Atlantic basin favors stability of the NADW cell by inhibiting the AAIW reverse cell.

Fig. 7.

Time series of (a) NADW production rate and (b) the AAIW reverse cell for experiments CNTRL under a perturbation attaining a maximum value of 0.11 Sv (0.38 m yr−1, solid) and experiment ⅓KVAtl under perturbations attaining a maximum of 0.14 (0.51 m yr−1, dashed), 0.29 (1.02 m yr−1, dashed–dotted), and 0.43 Sv (1.53 m yr−1, dotted). Values are given in Sv.

Fig. 7.

Time series of (a) NADW production rate and (b) the AAIW reverse cell for experiments CNTRL under a perturbation attaining a maximum value of 0.11 Sv (0.38 m yr−1, solid) and experiment ⅓KVAtl under perturbations attaining a maximum of 0.14 (0.51 m yr−1, dashed), 0.29 (1.02 m yr−1, dashed–dotted), and 0.43 Sv (1.53 m yr−1, dotted). Values are given in Sv.

The enhanced stability of the NADW cell in ⅓KVAtl is not the result of changes in surface density or isotherm depth. First, the surface conditions in ⅓KVAtl indicate a cooling and freshening of the NADW formation regions relative to CNTRL (figure not shown), with the net effect of this change a reduction of surface density in the North Atlantic. This proves that changes in NA surface density in ⅓KVAtl are not responsible for the increased stability of NADW. Second, as expected, we observe a shoaling of the isotherms in the upper 2000 m of the Atlantic in ⅓KVAtl relative to CNTRL (figure not shown). A deepening of Atlantic isotherms is known to enhance NADW formation (Gnanadesikan 1999) and perhaps its stability, yet we find shoaling of the Atlantic isotherms in the upper 2000 m in ⅓KVAtl. Moreover, the slightly reduced rate of NADW formation in ⅓KVAtl (19.8 Sv) might simplistically suggest reduced NADW stability, the opposite to what we have found. We have shown in contrast that a reduction in vertical mixing inside the Atlantic leads to an increased stability of NADW formation. This results from the weakening of the AAIW reverse cell caused by the reduced vertical mixing in the Atlantic sector.

A very different picture emerges when Kυ is reduced in the Indian and Pacific Oceans. We have tried to excite a transition to an NADW on state in ⅓KVIP by applying FW extractions from the NA without success. For example, Fig. 8 shows the time series for experiment ⅓KVIP under an FW extraction attaining a maximum of 1.7 m yr−1 after 150 yr. Time series are shown for the NADW production rate and the AAIW reverse cell clearly showing the inability of the ⅓KVIP case to maintain stable NADW production despite a strong FW extraction. Apparently, in our model, the Indian and Pacific Oceans comprise an area of significant deepwater removal by vertical mixing. This sensitivity of NADW to vertical mixing inside the Indian and Pacific Oceans indicates that in our model, alternative removal mechanisms, such as mechanically driven upwelling due to wind stress in the Southern Ocean, are not sufficient alone to maintain the NADW cell in the absence of strong Indo-Pacific vertical mixing. A dye tracer experiment tagging NADW recirculation (figure not shown) shows upwelling of NADW in the lower latitudes of the Indian and Pacific Oceans, as well as at some locations in the high latitudes of the Southern Ocean. This confirms the importance of the Indian/Pacific and Southern Oceans in the removal of NADW from the deep oceans in this model. In this case, a reduction of vertical mixing inside the Indian and Pacific Oceans shifts the competitive advantage of the AAIW/NADW cells in favor of the AAIW cell. We have also run an experiment reducing Kυ globally by a factor of 1/3 and again find weak overturning in the Atlantic basin. In this case, both the AAIW reverse cell and the NADW cell are suppressed.

Fig. 8.

Time series for experiment ⅓KVIP under the FW perturbation shown in Fig. 2 attaining a minimum M of −0.48 Sv (−1.7 m yr−1) after 150 yr. Time series are shown for (a) NADW production rate and (b) the AAIW reverse cell. Values are given in Sv.

Fig. 8.

Time series for experiment ⅓KVIP under the FW perturbation shown in Fig. 2 attaining a minimum M of −0.48 Sv (−1.7 m yr−1) after 150 yr. Time series are shown for (a) NADW production rate and (b) the AAIW reverse cell. Values are given in Sv.

Figure 9 shows a time series of the CNTRL experiment under an FW pulse attaining a maximum value M of 0.1 Sv (0.34 m yr−1). A recovery of NADW formation occurs under this perturbation. As in the experiments of Gregory et al. (2003), the recovery of the NADW cell is simultaneous with the disappearance of the AAIW reverse cell. This is in contrast to the experiments ¼KVAtl and ⅓KVAtl shown in Fig. 6, where NADW formation recovers some time after the AAIW reverse cell has disappeared. This suggests that reducing Kυ inside the Atlantic also removes the synchroneity between the reestablishment of the NADW cell and the disappearance of the AAIW reverse cell in ocean models.

c. Importance of location of change in Kυ

Figure 10 shows time series of NADW production rate and the AAIW reverse cell for experiment MS99KVAtl under FW perturbations attaining a maximum of 0.08 and 0.11 Sv and for experiment MS99KVglob under FW perturbations attaining a maximum of 0.11 and 0.76 Sv. In agreement with MS99 we find a significantly enhanced NADW stability in MS99KVglob, where the 0.76-Sv perturbation is not sufficient to permanently shut down NADW formation. This should be compared to the much lower value of 0.11 Sv that is sufficient to excite a transition to a stable NADW off state in CNTRL (Fig. 7). The NADW collapse in MS99KVAtl under a perturbation of only 0.08 Sv (Fig. 10), however, shows that increasing Kυ only inside the Atlantic results in a reduction of NADW stability compared to the enhanced stability found under a global increase in Kυ. The reduction in NADW stability in MS99KVAtl arises from the significant increase in strength of the AAIW reverse cell in response to increased vertical mixing inside the Atlantic. This is consistent with our previous findings in experiments ⅓KVAtl and ⅔KVAtl. In contrast, we could not obtain a stable off state for MS99KVglob. In summary, a markedly different response of the ocean’s MOC is obtained when altering Kυ globally as compared to Kυ changes limited to the Atlantic Ocean.

Fig. 10.

Time series of (a) NADW production rate and (b) the AAIW reverse cell strength for experiment MS99KVglob and MS99KVAtl under several FW perturbation scenarios (see legend for details). Values are given in Sv.

Fig. 10.

Time series of (a) NADW production rate and (b) the AAIW reverse cell strength for experiment MS99KVglob and MS99KVAtl under several FW perturbation scenarios (see legend for details). Values are given in Sv.

4. Summary and conclusions

In this article we have demonstrated that regional variations in vertical mixing can fundamentally affect the global meridional overturning circulation and its stability to FW perturbations. An important reason for applying such changes in Kυ in our model is to demonstrate the role of the AAIW reverse cell in maintaining the stability of circulation states. Our results show that a reduction of Kυ inside the Atlantic basin can drastically increase NADW stability with respect to FW perturbations applied to the NA. Conversely, a reduction of Kυ inside the Indian and Pacific basins can inhibit NADW formation, enabling the establishment of an AAIW reverse cell driven by upwelling inside the Atlantic basin. Furthermore, the FW perturbations we apply are unable to excite transitions to a stable NADW “off” state for the cases where Kυ is multiplied by a factor of 1/2 or less in the Atlantic Ocean. Multiplication of Kυ in the Atlantic by a factor of 2/3, however, allows the model to admit a transition to a stable NADW off state in response to the FW perturbations. The reduction in strength and depth of the AAIW reverse cell in this NADW off state illustrates that the AAIW reverse cell is indeed inhibited by a reduction of Kυ in the Atlantic Ocean. It should be noted that we have rerun a version of ⅓KVAtl using the computationally more intensive Gent and McWilliams (1990) eddy advection scheme combined with along-isopycnal mixing and no horizontal diffusion and also found increased NADW stability in this experiment.

Because a reduction in vertical mixing in the Atlantic inhibits the AAIW reverse cell, it in turn enhances the stability of the NADW cell. Indeed, Saenko et al. (2003) and Gregory et al. (2003) suggest that the AAIW reverse cell is responsible for the stability of the NADW off state, which we have also demonstrated in this study. After an initial suppression of NADW due to freshening of the NADW formation regions by an FW pulse (Fig. 6), no stable AAIW reverse cell develops when Kυ is reduced significantly in the Atlantic. The recovery of NADW formation in the absence of this cell in ⅓KVAtl supports the idea that the AAIW reverse cell (associated with NADW off states in ocean models) is a contributing factor to the stability of collapsed NADW states. Furthermore, changes in surface conditions and pycnocline depth inside the Atlantic do not contribute to the enhanced NADW stability we find in ⅓KVAtl. Rather, it appears to be exclusively due to the suppression of the AAIW reverse cell in the Southern Hemisphere in this experiment.

Although detrimental to the AAIW reverse cell, reduced Kυ inside the Atlantic does not inhibit the NADW cell, as its deepwater removal branch is located elsewhere, in the Indian, Pacific, and Southern Oceans. Our inability to obtain a stable NADW “on” state in experiment ⅓KVIP shows that the Indian and Pacific basins are important locations for the removal of NADW in our model. Indeed, reduced vertical mixing inside the Indian and Pacific basins in ⅓KVIP inhibits the eventual removal of NADW from the Atlantic basin and is therefore detrimental to NADW formation. In contrast, the similarity between the Atlantic MOC of the NADW off state in CNTRL and ⅓KVIP shows that the AAIW reverse cell is not affected by the reduction of vertical mixing in the Indian and Pacific Oceans. This is not surprising as its upwelling branch occurs inside the Atlantic basin. Therefore, a reduction of Kυ inside the Indian and Pacific Oceans shifts the competitive advantage from the NADW cell in favor of the AAIW reverse cell. While not explored here, this may have implications for ancient states of the global THC, as the distribution of seafloor “roughness” has evolved gradually over geological time scales. In particular, changes in the Southern Ocean could play an important role.

A global increase of Kυ in our model results in significantly enhanced NADW stability. However, an identical enhancement of Kυ only inside the Atlantic Ocean results in a reduction in NADW stability with respect to the control experiment CNTRL. Our results are thus not in contradiction with the enhanced NADW stability under globally increased Kυ found by MS99. However, we find a markedly different response when Kυ is increased only within the Atlantic Ocean, as in that case NADW stability is diminished. This suggests that the enhanced stability of the NADW on state found by MS99 under globally increased Kυ is due to the increase in vertical mixing outside the Atlantic Ocean, where NADW resurfaces.

The polarity of the global MOC is determined by interhemispheric competition between the AAIW reverse cell and the NADW cell (in today’s climate the AABW cell plays a minor role due to the Drake Passage effect; Sijp and England 2005). Unlike the idealized symmetric competition between an SH cell and an NH cell described by Bryan (1986), the NADW and AAIW cells depend on upwelling at different locations. In contrast to the AAIW reverse cell, the NADW cell depends on removal of its deepwater in areas outside the Atlantic, including the Southern and low-latitude Indian/Pacific Oceans. While vertical mixing plays an important role in NADW removal in our results, wind-driven upwelling in the Southern Ocean is thought to be another important mechanism to remove NADW from the deep ocean (Toggweiler and Samuels 1995). NADW is sensitive to vertical mixing in the Indian and Pacific Oceans in our model as a substantial component of NADW upwells into those sectors in the simulations (as in Gordon 1986). If wind-driven upwelling was a more prominent NADW removal mechanism in our model, perhaps by employing vertical mixing schemes with lower background diffusivity values, we might expect a reduction in this sensitivity. For instance, Saenko and Merryfield (2005) employ a parameterization of tidally driven deep mixing combining mixing near rough topography with a low background diffusivity of 10−1 cm2 s−1. As this tidally driven mixing over rough topography mainly occurs below the pycnocline, vertical mixing inside the pycnocline is small. In this model setup, Saenko and Merryfield (2005) find a dominance of Southern Ocean deepwater removal, over the classical notion of uniform abyssal upwelling (Munk 1966; Stommel and Arons 1960), and the conveyor pathway schematic of Gordon (1986). We would nonetheless expect the sensitivity of the AAIW reverse cell to vertical mixing in the Atlantic to be robust in experiments wherein wind-driven upwelling in the Southern Ocean dominates the removal of NADW.

We have shown that the relative strength of the NADW/AAIW cells, and therefore NADW stability, depends not only on surface density at the NADW and AAIW formation regions, but also on the relative strength of their removal mechanisms. This removal takes place in different geographic locations, so that the relative strengths of the NADW/AAIW cells can be altered by changing the vertical mixing coefficient in these regions. In the real ocean, NADW recirculation depends on vertical mixing across the thermocline and mechanical wind-driven upwelling in the Southern Ocean. Of these fundamentally different processes, vertical mixing appears to be regulating the removal of deepwater when the AAIW reverse cell operates in a NADW off state. This has implications for the understanding of the global thermohaline circulation in past and future climates.

Acknowledgments

We thank the University of Victoria staff for support in usage of their coupled climate model. We thank Jonathan Gregory, Ron Stouffer, and an anonymous reviewer for useful comments and suggestions that allowed us to greatly enhance the clarity and content of this study. This research was supported by the Australian Research Council and the Australian Antarctic Science Program.

REFERENCES

REFERENCES
Broecker
,
W. S.
,
1991
:
The great ocean conveyor belt.
Oceanography
,
4
,
79
87
.
Bryan
,
F.
,
1986
:
High-latitude salinity effects and interhemispheric thermohaline circulations.
Nature
,
323
,
301
304
.
Bryan
,
F.
,
1987
:
Parameter sensitivity of primitive equation ocean general circulation models.
J. Phys. Oceanogr.
,
17
,
970
985
.
Döös
,
K.
,
1995
:
Inter-ocean exchange of water masses.
J. Geophys. Res.
,
100
,
13499
13514
.
Duffy
,
P. B.
, and
K.
Caldeira
,
1997
:
Sensitivity of simulated salinity in a three dimensional ocean model to upper-ocean transport of salt from sea-ice formation.
Geophys. Res. Lett.
,
24
,
1323
1326
.
Gent
,
P. R.
, and
J. C.
McWilliams
,
1990
:
Isopycnal mixing in ocean general circulation models.
J. Phys. Oceanogr.
,
20
,
150
155
.
Gnanadesikan
,
A.
,
1999
:
A simple predictive model for the structure of the oceanic pycnocline.
Science
,
283
,
2077
2079
.
Gordon
,
A. L.
,
1986
:
Interocean exchange of thermocline water.
J. Geophys. Res.
,
91
,
5037
5046
.
Gregory
,
J. M.
,
O. A.
Saenko
, and
A. J.
Weaver
,
2003
:
The role of the Atlantic freshwater balance in the hyteresis of the meridional overturning circulation.
Climate Dyn.
,
21
.
doi:10.1007/s00382-003-0359-8
.
Hasumi
,
H.
, and
N.
Suginohara
,
1999
:
Effects of locally enhanced vertical diffusivity over rough bathymetry on the world ocean circulation.
J. Geophys. Res.
,
104
,
C10
.
23367
23374
.
Kalnay
,
E.
, and
Coauthors
,
1996
:
The NCEP/NCAR 40-Year Reanalysis Project.
Bull. Amer. Meteor. Soc.
,
77
,
437
471
.
Ledwell
,
J.
,
E.
Montgomery
,
K.
Polzin
,
L. S.
Laurent
,
R.
Schmitt
, and
J.
Toole
,
2000
:
Evidence for enhanced mixing over rough topography in the abyssal ocean.
Nature
,
403
,
179
182
.
Luyten
,
J. R.
,
J.
Pedlosky
, and
H.
Stommel
,
1983
:
The ventilated thermocline.
J. Phys. Oceanogr.
,
13
,
292
309
.
Manabe
,
S.
, and
R. J.
Stouffer
,
1988
:
Two stable equilibria of a coupled ocean–atmosphere model.
J. Climate
,
1
,
841
866
.
Manabe
,
S.
, and
R. J.
Stouffer
,
1999
:
Are two modes of thermohaline circulation stable?
Tellus
,
51A
,
400
411
.
Mikolajewicz
,
U.
,
E.
Maier-Reimer
,
T. J.
Crowley
, and
K. Y.
Kim
,
1993
:
Effect of Drake Passage and Panamanian gateways on the circulation of an ocean model.
Paleoceanography
,
8
,
409
426
.
Munk
,
W. H.
,
1966
:
Abyssal recipes.
Deep-Sea Res.
,
13
,
707
730
.
Munk
,
W. H.
, and
C.
Wunsch
,
1998
:
Abyssal recipes II: Energetics of tidal and wind mixing.
Deep-Sea Res.
,
45
,
1977
2010
.
Pacanowski
,
R.
,
1995
:
MOM2 documentation user’s guide and reference manual: GFDL ocean group technical report. NOAA/GFDL, Princeton, NJ, 232 pp
.
Prange
,
P.
,
G.
Lohmann
, and
A.
Paul
,
2003
:
Influence of vertical mixing on the thermohaline hysteresis: Analysis of an OGCM.
J. Phys. Oceanogr.
,
33
,
1707
1719
.
Rahmstorf
,
S.
,
1996
:
On the freshwater forcing and transport of the Atlantic thermohaline circulation.
Climate Dyn.
,
12
,
799
811
.
Redi
,
M. H.
,
1982
:
Oceanic isopycnal mixing by coordinate rotation.
J. Phys. Oceanogr.
,
12
,
1154
1158
.
Saenko
,
O. A.
, and
W. J.
Merryfield
,
2005
:
On the effect of topographically enhanced mixing on the global ocean circulation.
J. Phys. Oceanogr.
,
35
,
826
834
.
Saenko
,
O. A.
,
A. J.
Weaver
, and
J. M.
Gregory
,
2003
:
On the link between the two modes of the ocean thermohaline circulation and the formation of global-scale water masses.
J. Climate
,
16
,
2797
2801
.
Schmittner
,
A.
, and
A. J.
Weaver
,
2001
:
Dependence of multiple climate states on ocean mixing parameters.
Geophys. Res. Lett.
,
28
,
1027
1030
.
Sijp
,
W. P.
, and
M. H.
England
,
2004
:
Effect of the Drake Passage throughflow on global climate.
J. Phys. Oceanogr.
,
34
,
1254
1266
.
Sijp
,
W. P.
, and
M. H.
England
,
2005
:
On the role of the Drake Passage in controlling the stability of the ocean’s thermohaline circulation.
J. Climate
,
18
,
1957
1966
.
Stommel
,
H.
,
1961
:
Thermohaline convection with two stable regimes of flow.
Tellus
,
13
,
224
230
.
Stommel
,
H.
, and
A. B.
Arons
,
1960
:
On the abyssal circulation of the world ocean, I. An idealized model of the circulation pattern and amplitude in oceanic basins.
Deep-Sea Res.
,
6
,
217
233
.
Toggweiler
,
J. R.
, and
B. L.
Samuels
,
1995
:
Effect of Drake Passage on the global thermohaline circulation.
Deep-Sea Res. I
,
42
,
477
500
.
Weaver
,
A. J.
, and
Coauthors
,
2001
:
The UVic Earth System Climate Model: Model description, climatology, and applications to past, present and future climates.
Atmos.–Ocean
,
39
,
1067
1109
.

Footnotes

Corresponding author address: Willem P. Sijp, School of Mathematics and Statistics, University of New South Wales, Sydney, NSW 2052, Australia. Email: wsijp@maths.unsw.edu.au

1

NADW outflow is measured as the local maximum at intermediate depth of the meridional transport streamfunction at 30°S.

2

We calculate this recirculation by subtracting the NADW outflow rate from the NADW formation rate.