Search Results
You are looking at 11 - 20 of 26 items for :
- Author or Editor: Alexey Fedorov x
- Article x
- Refine by Access: Content accessible to me x
Abstract
A model of surface waves generated on deep water by strong winds is proposed. A two-layer approximation is adopted, in which a shallow turbulent layer overlies the lower, infinitely deep layer. The dynamics of the upper layer, which is directly exposed to the wind, are nonlinear and coupled to the linear dynamics in the deep fluid. The authors demonstrate that in such a system there exist steady wave solutions characterized by confined regions of wave breaking alternating with relatively long intervals where the wave profiles change monotonically. In the former regions the flow is decelerated; in the latter it is accelerated. The regions of breaking are akin to hydraulic jumps of finite width necessary to join the smooth “interior” flows and have periodic waves. In contrast to classical hydraulic jumps, the strongly forced waves lose both energy and momentum across the jumps. The flow in the upper layer is driven by the balance between the wind stress at the surface, the turbulent drag applied at the layer interface, and the wave drag induced at the layer interface by quasi-steady breaking waves. Propagating in the downwind direction, the strongly forced waves significantly modify the flow in both layers, lead to enhanced turbulence, and reduce the speed of the near-surface flow. According to this model, a large fraction of the work done by the surface wind stress on the ocean in high winds may go directly into wave breaking and surface turbulence.
Abstract
A model of surface waves generated on deep water by strong winds is proposed. A two-layer approximation is adopted, in which a shallow turbulent layer overlies the lower, infinitely deep layer. The dynamics of the upper layer, which is directly exposed to the wind, are nonlinear and coupled to the linear dynamics in the deep fluid. The authors demonstrate that in such a system there exist steady wave solutions characterized by confined regions of wave breaking alternating with relatively long intervals where the wave profiles change monotonically. In the former regions the flow is decelerated; in the latter it is accelerated. The regions of breaking are akin to hydraulic jumps of finite width necessary to join the smooth “interior” flows and have periodic waves. In contrast to classical hydraulic jumps, the strongly forced waves lose both energy and momentum across the jumps. The flow in the upper layer is driven by the balance between the wind stress at the surface, the turbulent drag applied at the layer interface, and the wave drag induced at the layer interface by quasi-steady breaking waves. Propagating in the downwind direction, the strongly forced waves significantly modify the flow in both layers, lead to enhanced turbulence, and reduce the speed of the near-surface flow. According to this model, a large fraction of the work done by the surface wind stress on the ocean in high winds may go directly into wave breaking and surface turbulence.
Abstract
Global climate models frequently exhibit cold biases in tropical sea surface temperature (SST) in the central and eastern equatorial Pacific. Here, Lagrangian particle back trajectories are used to investigate the source regions of the water that upwells along the equator in the IPSL climate model to test and confirm the hypothesis that the SST biases are caused by remote biases advected in from the extratropics and to identify the dominant source regions. Water in the model is found to be sourced primarily from localized regions along the western and eastern flanks of the subtropical gyres. However, while the model SST bias is especially large in the northwestern subtropical Pacific (about −5°C), it is found that the eastern subtropics contribute to the equatorial bias the most. This is due to two distinct subsurface pathways connecting these regions to the equator. The first pathway, originating in the northwestern subtropical Pacific, has relatively long advection time scales close to or exceeding 60 yr, wherein particles recirculate around the subtropical gyres while descending to approximately 500 m before then shoaling toward the equatorial undercurrent. The second pathway, from the eastern subtropics, has time scales close to 10 yr, with particles following a shallow and more direct route to the equator within the upper 200 m. The deeper and longer pathway taken by the western subtropical water ensures that vertical mixing can erode the bias. Ultimately, it is estimated that relatively confined regions in the eastern subtropics of both hemispheres control approximately half of the equatorial bias.
Abstract
Global climate models frequently exhibit cold biases in tropical sea surface temperature (SST) in the central and eastern equatorial Pacific. Here, Lagrangian particle back trajectories are used to investigate the source regions of the water that upwells along the equator in the IPSL climate model to test and confirm the hypothesis that the SST biases are caused by remote biases advected in from the extratropics and to identify the dominant source regions. Water in the model is found to be sourced primarily from localized regions along the western and eastern flanks of the subtropical gyres. However, while the model SST bias is especially large in the northwestern subtropical Pacific (about −5°C), it is found that the eastern subtropics contribute to the equatorial bias the most. This is due to two distinct subsurface pathways connecting these regions to the equator. The first pathway, originating in the northwestern subtropical Pacific, has relatively long advection time scales close to or exceeding 60 yr, wherein particles recirculate around the subtropical gyres while descending to approximately 500 m before then shoaling toward the equatorial undercurrent. The second pathway, from the eastern subtropics, has time scales close to 10 yr, with particles following a shallow and more direct route to the equator within the upper 200 m. The deeper and longer pathway taken by the western subtropical water ensures that vertical mixing can erode the bias. Ultimately, it is estimated that relatively confined regions in the eastern subtropics of both hemispheres control approximately half of the equatorial bias.
Abstract
This study compares the impacts of Arctic sea ice decline on the Atlantic meridional overturning circulation (AMOC) in two configurations of the Community Earth System Model with different horizontal resolution. In a suite of model experiments, we impose radiative imbalance at the ice surface, replicating a loss of sea ice cover comparable to that observed during 1979–2014, and we find dramatic differences in the AMOC response between the two models. In the lower-resolution configuration, the AMOC weakens by about one-third over the first 100 years, approaching a new quasi-equilibrium. By contrast, in the higher-resolution configuration, the AMOC weakens by ~10% during the first 20–30 years followed by a full recovery driven by invigorated deep water formation in the Labrador Sea and adjacent regions. We investigate these differences using a diagnostic AMOC stability indicator, which reflects the AMOC freshwater transport in and out of the basin and hence the strength of the basin-scale salt-advection feedback. This indicator suggests that the AMOC in the lower-resolution model is less stable and more sensitive to surface perturbations, as confirmed by hosing experiments mimicking Arctic freshening due to sea ice decline. Differences between the models’ mean states, including the Atlantic Ocean mean surface freshwater fluxes, control the differences in AMOC stability. Our results demonstrate that the AMOC stability indicator is indeed useful for evaluating AMOC sensitivity to perturbations. We emphasize that, despite the differences in the long-term adjustment, both models simulate a multidecadal AMOC weakening caused by Arctic sea ice decline, relevant to climate change.
Abstract
This study compares the impacts of Arctic sea ice decline on the Atlantic meridional overturning circulation (AMOC) in two configurations of the Community Earth System Model with different horizontal resolution. In a suite of model experiments, we impose radiative imbalance at the ice surface, replicating a loss of sea ice cover comparable to that observed during 1979–2014, and we find dramatic differences in the AMOC response between the two models. In the lower-resolution configuration, the AMOC weakens by about one-third over the first 100 years, approaching a new quasi-equilibrium. By contrast, in the higher-resolution configuration, the AMOC weakens by ~10% during the first 20–30 years followed by a full recovery driven by invigorated deep water formation in the Labrador Sea and adjacent regions. We investigate these differences using a diagnostic AMOC stability indicator, which reflects the AMOC freshwater transport in and out of the basin and hence the strength of the basin-scale salt-advection feedback. This indicator suggests that the AMOC in the lower-resolution model is less stable and more sensitive to surface perturbations, as confirmed by hosing experiments mimicking Arctic freshening due to sea ice decline. Differences between the models’ mean states, including the Atlantic Ocean mean surface freshwater fluxes, control the differences in AMOC stability. Our results demonstrate that the AMOC stability indicator is indeed useful for evaluating AMOC sensitivity to perturbations. We emphasize that, despite the differences in the long-term adjustment, both models simulate a multidecadal AMOC weakening caused by Arctic sea ice decline, relevant to climate change.
Abstract
The behavior of rotating and nonrotating aggregated convection is examined at various horizontal resolutions using the hypohydrostatic, or reduced acceleration in the vertical (RAVE), rescaling. This modification of the equations of motion reduces the scale separation between convective- and larger-scale motions, enabling the simultaneous and explicit representation of both types of flow in a single model without convective parameterization. Without the RAVE rescaling, a dry bias develops when simulations of nonrotating radiative–convective equilibrium are integrated at coarse resolution in domains large enough to permit convective self-aggregation. The rescaling reduces this dry bias, and here it is suggested that the rescaling moistens the troposphere by weakening the amplitude and slowing the group velocity of gravity waves, thus reducing the subsidence drying around aggregated convection. Separate simulations of rotating radiative–convective equilibrium exhibit tropical cyclogenesis; as horizontal resolution is coarsened without the rescaling, the resulting storms intensify more slowly and achieve lower peak intensities. At a given horizontal resolution, using RAVE increases peak storm intensity and reduces the time needed for tropical cyclogenesis—effects here suggested to be caused at least in part by the environmental moistening produced by RAVE. Consequently, the RAVE rescaling has the potential to improve simulations of tropical cyclones and other aggregated convection in models with horizontal resolutions of order 10–100 km.
Abstract
The behavior of rotating and nonrotating aggregated convection is examined at various horizontal resolutions using the hypohydrostatic, or reduced acceleration in the vertical (RAVE), rescaling. This modification of the equations of motion reduces the scale separation between convective- and larger-scale motions, enabling the simultaneous and explicit representation of both types of flow in a single model without convective parameterization. Without the RAVE rescaling, a dry bias develops when simulations of nonrotating radiative–convective equilibrium are integrated at coarse resolution in domains large enough to permit convective self-aggregation. The rescaling reduces this dry bias, and here it is suggested that the rescaling moistens the troposphere by weakening the amplitude and slowing the group velocity of gravity waves, thus reducing the subsidence drying around aggregated convection. Separate simulations of rotating radiative–convective equilibrium exhibit tropical cyclogenesis; as horizontal resolution is coarsened without the rescaling, the resulting storms intensify more slowly and achieve lower peak intensities. At a given horizontal resolution, using RAVE increases peak storm intensity and reduces the time needed for tropical cyclogenesis—effects here suggested to be caused at least in part by the environmental moistening produced by RAVE. Consequently, the RAVE rescaling has the potential to improve simulations of tropical cyclones and other aggregated convection in models with horizontal resolutions of order 10–100 km.
Abstract
Interactions between the tropical oceans and atmosphere permit a spectrum of natural modes of oscillation whose properties—period, intensity, spatial structure, and direction of propagation—depend on the background climatic state (i.e., the mean state). This mean state can be described by parameters that include the following: the time-averaged intensity τ of the Pacific trade winds, the mean depth (H) of the thermocline, and the temperature difference across the thermocline (ΔT). A stability analysis by means of a simple coupled ocean–atmosphere model indicates two distinct families of unstable modes. One has long periods of several years, involves sea surface temperature variations determined by vertical movements of the thermocline that are part of the adjustment of the ocean basin to the fluctuating winds, requires a relatively deep thermocline, and corresponds to the delayed oscillator. The other family requires a shallow thermocline, has short periods of a year or two, has sea surface temperature variations determined by advection and by entrainment across the thermocline, and is associated with westward phase propagation. For the modes to be unstable, both families require that the background zonal wind exceed a certain intensity. An increase in ΔT, and in H beyond a certain value, are stabilizing. For intermediate values of H, between large values that favor the one mode and small values that favor the other, the modes are of a hybrid type with some properties of each family. The observed Southern Oscillation has been of this type for the past few decades, but some paleorecords suggest that, in the distant past, the oscillation was strictly of the delayed oscillator type and had a very long period on the order of a decade.
Abstract
Interactions between the tropical oceans and atmosphere permit a spectrum of natural modes of oscillation whose properties—period, intensity, spatial structure, and direction of propagation—depend on the background climatic state (i.e., the mean state). This mean state can be described by parameters that include the following: the time-averaged intensity τ of the Pacific trade winds, the mean depth (H) of the thermocline, and the temperature difference across the thermocline (ΔT). A stability analysis by means of a simple coupled ocean–atmosphere model indicates two distinct families of unstable modes. One has long periods of several years, involves sea surface temperature variations determined by vertical movements of the thermocline that are part of the adjustment of the ocean basin to the fluctuating winds, requires a relatively deep thermocline, and corresponds to the delayed oscillator. The other family requires a shallow thermocline, has short periods of a year or two, has sea surface temperature variations determined by advection and by entrainment across the thermocline, and is associated with westward phase propagation. For the modes to be unstable, both families require that the background zonal wind exceed a certain intensity. An increase in ΔT, and in H beyond a certain value, are stabilizing. For intermediate values of H, between large values that favor the one mode and small values that favor the other, the modes are of a hybrid type with some properties of each family. The observed Southern Oscillation has been of this type for the past few decades, but some paleorecords suggest that, in the distant past, the oscillation was strictly of the delayed oscillator type and had a very long period on the order of a decade.
Abstract
Properties of internal wave fronts or Kelvin fronts travelling eastward in the equatorial waveguide are studied, motivated by recent studies on coastal Kelvin waves and jumps and new data on equatorial Kelvin waves. It has been recognized for some time that nonlinear equatorial Kelvin waves can steepen and break, forming a broken wave of depression, or front, propagating eastward. The three-dimensional structure of the wave field associated with such a front is considered. As for linear Kelvin waves, the front is symmetrical with respect to the equator. Sufficiently far away from the front, the wave profile is Gaussian in the meridional direction, with the equatorial Rossby radius of deformation being its decay scale. Due to nonlinearity, the phase speed of the front is greater than that of linear Kelvin waves, resulting in a supercritical flow. This leads to the resonant generation of equatorially trapped gravity–inertial (or Poincaré) waves, analogous in principle to the resonant mechanism for nonlinear coastal Kelvin waves. First-mode symmetrical Poincaré waves are generated, with their wavelength determined by the amplitude of the front. Finally, the propagation of a Kelvin front gives rise to a nonzero poleward mass transport above the thermocline, in consequence of which there is a poleward heat flux.
Abstract
Properties of internal wave fronts or Kelvin fronts travelling eastward in the equatorial waveguide are studied, motivated by recent studies on coastal Kelvin waves and jumps and new data on equatorial Kelvin waves. It has been recognized for some time that nonlinear equatorial Kelvin waves can steepen and break, forming a broken wave of depression, or front, propagating eastward. The three-dimensional structure of the wave field associated with such a front is considered. As for linear Kelvin waves, the front is symmetrical with respect to the equator. Sufficiently far away from the front, the wave profile is Gaussian in the meridional direction, with the equatorial Rossby radius of deformation being its decay scale. Due to nonlinearity, the phase speed of the front is greater than that of linear Kelvin waves, resulting in a supercritical flow. This leads to the resonant generation of equatorially trapped gravity–inertial (or Poincaré) waves, analogous in principle to the resonant mechanism for nonlinear coastal Kelvin waves. First-mode symmetrical Poincaré waves are generated, with their wavelength determined by the amplitude of the front. Finally, the propagation of a Kelvin front gives rise to a nonzero poleward mass transport above the thermocline, in consequence of which there is a poleward heat flux.
Abstract
The evolution of nonlinear Kelvin waves is studied using analytical and numerical methods. In the absence of dispersive (nonhydrostatic) effects, such waves may evolve to braking. The authors find that one of the effects of rotation is to delay the onset of breaking in time by up to 60%, with respect to a comparable wave in de absence of rotation. This delay is consistent with qualitative conclusions based on transverse averaging of the evolution equations. Further, the onset of breaking occurs almost simultaneously over a zone of uniform phase that is normal to the boundary and extends over a distance comparable to the Rossby radius of deformation. In other words, the process of breaking embraces the most energetic area of the wave. In contrast to the linear Kelvin wave, the nonlinear wave develops a dipole structure in the cross-shelf velocity, with a zero net offshore flow. With increasing nonlinearity the flow develops a stronger offshore jet ahead of the wave crest. The Kelvin wave amplitude at the coast delays slightly with time. This and other major features of the wave are accounted for by an analytical model based on slowly varying averaged variables. As part of the analysis it is demonstrated that the evolution of the wave phase may be described by an inhomogeneous Klein-Gordon equation.
Abstract
The evolution of nonlinear Kelvin waves is studied using analytical and numerical methods. In the absence of dispersive (nonhydrostatic) effects, such waves may evolve to braking. The authors find that one of the effects of rotation is to delay the onset of breaking in time by up to 60%, with respect to a comparable wave in de absence of rotation. This delay is consistent with qualitative conclusions based on transverse averaging of the evolution equations. Further, the onset of breaking occurs almost simultaneously over a zone of uniform phase that is normal to the boundary and extends over a distance comparable to the Rossby radius of deformation. In other words, the process of breaking embraces the most energetic area of the wave. In contrast to the linear Kelvin wave, the nonlinear wave develops a dipole structure in the cross-shelf velocity, with a zero net offshore flow. With increasing nonlinearity the flow develops a stronger offshore jet ahead of the wave crest. The Kelvin wave amplitude at the coast delays slightly with time. This and other major features of the wave are accounted for by an analytical model based on slowly varying averaged variables. As part of the analysis it is demonstrated that the evolution of the wave phase may be described by an inhomogeneous Klein-Gordon equation.
Abstract
El Niño–Southern Oscillation (ENSO) is a pronounced mode of climate variability that originates in the tropical Pacific and affects weather patterns worldwide. Growing evidence suggests that despite extensive changes in tropical climate, ENSO was active over vast geological epochs stretching millions of years from the late Cretaceous to the Holocene. In particular, ENSO persisted during the Pliocene, when a dramatic reduction occurred in the mean east–west temperature gradient in the equatorial Pacific. The mechanisms for sustained ENSO in such climates are poorly understood. Here a comprehensive climate model is used to simulate ENSO for a broad range of tropical Pacific mean climates characterized by different climatological SST gradients. It is found that the simulated ENSO remains surprisingly robust: when the east–west gradient is reduced from 6° to 1°C, the amplitude of ENSO decreases only by 30%–40%, its dominant period remains close to 3–4 yr, and the spectral peak stays above red noise. To explain these results, the magnitude of ocean–atmosphere feedbacks that control the stability of the natural mode of ENSO (the Bjerknes stability index) is evaluated. It is found that as a result of reorganization of the atmospheric Walker circulation in response to changes in the mean surface temperature gradient, the growth/decay rates of the ENSO mode stay nearly constant throughout different climates. These results explain the persistence of ENSO in the past and, in particular, reconcile the seemingly contradictory findings of ENSO occurrence and the small mean east–west temperature gradient during the Pliocene.
Abstract
El Niño–Southern Oscillation (ENSO) is a pronounced mode of climate variability that originates in the tropical Pacific and affects weather patterns worldwide. Growing evidence suggests that despite extensive changes in tropical climate, ENSO was active over vast geological epochs stretching millions of years from the late Cretaceous to the Holocene. In particular, ENSO persisted during the Pliocene, when a dramatic reduction occurred in the mean east–west temperature gradient in the equatorial Pacific. The mechanisms for sustained ENSO in such climates are poorly understood. Here a comprehensive climate model is used to simulate ENSO for a broad range of tropical Pacific mean climates characterized by different climatological SST gradients. It is found that the simulated ENSO remains surprisingly robust: when the east–west gradient is reduced from 6° to 1°C, the amplitude of ENSO decreases only by 30%–40%, its dominant period remains close to 3–4 yr, and the spectral peak stays above red noise. To explain these results, the magnitude of ocean–atmosphere feedbacks that control the stability of the natural mode of ENSO (the Bjerknes stability index) is evaluated. It is found that as a result of reorganization of the atmospheric Walker circulation in response to changes in the mean surface temperature gradient, the growth/decay rates of the ENSO mode stay nearly constant throughout different climates. These results explain the persistence of ENSO in the past and, in particular, reconcile the seemingly contradictory findings of ENSO occurrence and the small mean east–west temperature gradient during the Pliocene.
Abstract
The dynamics of El Niño–Southern Oscillation (ENSO) are studied in terms of the balance between energy input from the winds (via wind power) and changes in the storage of available potential energy in the tropical ocean. Presently, there are broad differences in the way global general circulation models simulate the dynamics, magnitude, and phase of ENSO events; hence, there is a need for simple, physically based metrics to allow for model evaluation. This energy description is a basinwide, integral, quantitative approach, ideal for intermodel comparison, that assesses model behavior in the subsurface ocean. Here it is applied to a range of ocean models and data assimilations within ENSO spatial and temporal scales. The onset of an El Niño is characterized by a decrease in wind power that leads to a decrease in available potential energy, and hence a flatter thermocline. In contrast, La Niña events are preceded by an increase in wind power that leads to an increase in the available potential energy and a steeper thermocline. The wind power alters the available potential energy via buoyancy power, associated with vertical mass fluxes that modify the slope of the isopycnals. Only a fraction of wind power is converted to buoyancy power. The efficiency of this conversion γ is estimated in this study at 50%–60%. Once the energy is delivered to the thermocline it is subject to small, but important, diffusive dissipation. It is estimated that this dissipation sets the e-folding damping rate α for the available potential energy on the order of 1 yr−1. The authors propose to use the efficiency γ and the damping rate α as two energy-based metrics for evaluating dissipative properties of the ocean component of general circulation models, providing a simple method for understanding subsurface ENSO dynamics and a diagnostic tool for exploring differences between the models.
Abstract
The dynamics of El Niño–Southern Oscillation (ENSO) are studied in terms of the balance between energy input from the winds (via wind power) and changes in the storage of available potential energy in the tropical ocean. Presently, there are broad differences in the way global general circulation models simulate the dynamics, magnitude, and phase of ENSO events; hence, there is a need for simple, physically based metrics to allow for model evaluation. This energy description is a basinwide, integral, quantitative approach, ideal for intermodel comparison, that assesses model behavior in the subsurface ocean. Here it is applied to a range of ocean models and data assimilations within ENSO spatial and temporal scales. The onset of an El Niño is characterized by a decrease in wind power that leads to a decrease in available potential energy, and hence a flatter thermocline. In contrast, La Niña events are preceded by an increase in wind power that leads to an increase in the available potential energy and a steeper thermocline. The wind power alters the available potential energy via buoyancy power, associated with vertical mass fluxes that modify the slope of the isopycnals. Only a fraction of wind power is converted to buoyancy power. The efficiency of this conversion γ is estimated in this study at 50%–60%. Once the energy is delivered to the thermocline it is subject to small, but important, diffusive dissipation. It is estimated that this dissipation sets the e-folding damping rate α for the available potential energy on the order of 1 yr−1. The authors propose to use the efficiency γ and the damping rate α as two energy-based metrics for evaluating dissipative properties of the ocean component of general circulation models, providing a simple method for understanding subsurface ENSO dynamics and a diagnostic tool for exploring differences between the models.
Abstract
Climate models show that most of the anthropogenic heat resulting from increased atmospheric CO2 enters the Southern Ocean near 60°S and is stored around 45°S. This heat is transported to the ocean interior by the meridional overturning circulation (MOC) with wind changes playing an important role in the process. To isolate and quantify the latter effect, we apply an overriding technique to a climate model and decompose the total ocean response to CO2 increase into two major components: one due to wind changes and the other due to direct CO2 effect. We find that the poleward-intensified zonal surface winds tend to shift and strengthen the ocean Deacon cell and hence the residual MOC, leading to anomalous divergence of ocean meridional heat transport around 60°S coupled to a surface heat flux increase. In contrast, at 45°S we see anomalous convergence of ocean heat transport and heat loss at the surface. As a result, the wind-induced ocean heat storage (OHS) peaks at 46°S at a rate of 0.07 ZJ yr−1 (° lat)−1 (1 ZJ = 1021 J), contributing 20% to the total OHS maximum. The direct CO2 effect, on the other hand, very slightly alters the residual MOC but primarily warms the ocean. It induces a small but nonnegligible change in eddy heat transport and causes OHS to peak at 42°S at a rate of 0.30 ZJ yr−1 (° lat)−1, accounting for 80% of the OHS maximum. We also find that the eddy-induced MOC weakens, primarily caused by a buoyancy flux change as a result of the direct CO2 effect, and does not compensate the intensified Deacon cell.
Abstract
Climate models show that most of the anthropogenic heat resulting from increased atmospheric CO2 enters the Southern Ocean near 60°S and is stored around 45°S. This heat is transported to the ocean interior by the meridional overturning circulation (MOC) with wind changes playing an important role in the process. To isolate and quantify the latter effect, we apply an overriding technique to a climate model and decompose the total ocean response to CO2 increase into two major components: one due to wind changes and the other due to direct CO2 effect. We find that the poleward-intensified zonal surface winds tend to shift and strengthen the ocean Deacon cell and hence the residual MOC, leading to anomalous divergence of ocean meridional heat transport around 60°S coupled to a surface heat flux increase. In contrast, at 45°S we see anomalous convergence of ocean heat transport and heat loss at the surface. As a result, the wind-induced ocean heat storage (OHS) peaks at 46°S at a rate of 0.07 ZJ yr−1 (° lat)−1 (1 ZJ = 1021 J), contributing 20% to the total OHS maximum. The direct CO2 effect, on the other hand, very slightly alters the residual MOC but primarily warms the ocean. It induces a small but nonnegligible change in eddy heat transport and causes OHS to peak at 42°S at a rate of 0.30 ZJ yr−1 (° lat)−1, accounting for 80% of the OHS maximum. We also find that the eddy-induced MOC weakens, primarily caused by a buoyancy flux change as a result of the direct CO2 effect, and does not compensate the intensified Deacon cell.
