We present results for simulated climate and climate change from a newly developed high-resolution global climate model (GFDL CM2.5). The GFDL CM2.5 model has an atmospheric resolution of approximately 50 Km in the horizontal, with 32 vertical levels. The horizontal resolution in the ocean ranges from 28 Km in the tropics to 8 Km at high latitudes, with 50 vertical levels. This resolution allows the explicit simulation of some mesoscale eddies in the ocean, particularly at lower latitudes.
We present analyses based on the output of a 280 year control simulation; we also present results based on a 140 year simulation in which atmospheric CO2 increases at 1% per year until doubling after 70 years.
Results are compared to the GFDL CM2.1 climate model, which has somewhat similar physics but coarser resolution. The simulated climate in CM2.5 shows marked improvement over many regions, especially the tropics, including a reduction in the double ITCZ and an improved simulation of ENSO. Regional precipitation features are much improved. The Indian monsoon and Amazonian rainfall are also substantially more realistic in CM2.5.
The response of CM2.5 to a doubling of atmospheric CO2 has many features in common with CM2.1, with some notable differences. For example, rainfall changes over the Mediterranean appear to be tightly linked to topography in CM2.5, in contrast to CM2.1 where the response is more spatially homogeneous. In addition, in CM2.5 the near-surface ocean warms substantially in the high latitudes of the Southern Ocean, in contrast to simulations using CM2.1.
Barreiro, M, Alexey Fedorov, Ronald C Pacanowski, and S G H Philander, 2008: Abrupt climate changes: How freshening of the northern Atlantic affects the thermohaline & wind-driven oceanic circulations. Annual Review of Earth and Planetary Sciences, 36, 33-58. Abstract
Leading hypotheses for abrupt climate changes are focused on the ocean response to a freshening of surface waters in the north Atlantic. The degree to which such a freshening affects the deep, slow thermohaline, rather than the shallow, swift, wind-driven circulations of the ocean, and hence the degree to which that freshening affects climate in high rather than low latitudes, differ from model to model, depending on factors such as the treatment of diffusive processes in the oceans. Many comprehensive climate models are biased and confine the influence mainly to the thermohaline circulation and northern climates. Simulations of paleoclimates can provide valuable tests for the models, but only some of those climates provide sufficiently stringent tests to determine which models are realistic.
Fedorov, Alexey, M Barreiro, G Boccaletti, Ronald C Pacanowski, and S G H Philander, April 2007: The freshening of surface waters in high latitudes: Effects on the thermohaline and wind-driven circulations. Journal of Physical Oceanography, 37(4), DOI:10.1175/JPO3033.1. Abstract
The impacts of a freshening of surface waters in high latitudes on the deep, slow, thermohaline circulation have received enormous attention, especially the possibility of a shutdown in the meridional overturning that involves sinking of surface waters in the northern Atlantic Ocean. A recent study by Federov,et al. has drawn attention to the effects of a freshening on the other main component of the oceanic circulation—the swift, shallow, wind-driven circulation that varies on decadal time scales and is closely associated with the ventilated thermocline. That circulation, too, involves meridional overturning, but its variations and critical transitions affect mainly the Tropics. A surface freshening in mid- to high latitudes can deepen the equatorial thermocline to such a degree that temperatures along the equator become as warm in the eastern part of the basin as they are in the west, the tropical zonal sea surface temperature gradient virtually disappears, and permanently warm conditions prevail in the Tropics. In a model that has both the wind driven and thermohaline components of the circulation, which factors determine the relative effects of a freshening on the two components and its impact on climate? Studies with an idealized ocean general circulation model find that vertical diffusivity is one of the critical parameters that affect the relative strength of the two circulation components and hence their response to a freshening. The spatial structure of the freshening and imposed meridional temperature gradients are other important factors.
During the early and mid-Pliocene, the period from 5 to 3 million years ago, approximately, the Earth is believed to have been significantly warmer than it is today, but the reasons for the higher temperatures are unclear. This paper explores the impact of recent findings that suggest that, at that time, cold surface waters were absent from the tropical and subtropical oceanic upwelling zones. El Niño was in effect a perennial rather than intermittent phenomenon, and sea surface temperatures in low latitudes were essentially independent of longitude. When these conditions are specified as the lower boundary condition for an atmospheric GCM, we find that the trade winds along the equator, and hence the Walker Circulation, collapse. The low-level stratus clouds in low latitudes diminish greatly, thus reducing the albedo of the Earth. The atmospheric concentration of water vapor increases, and enhanced latent heat release due to stronger evaporation warms up the tropical atmosphere, particularly between 40°S and 20°N. Moreover, teleconnection patterns from the Pacific induce a warming over North America that is enhanced by surface albedo feedback, a process that may have helped to maintain this region ice-free before 3 Ma. The results presented here indicate that the suggested absence of cold surface waters from the tropical and subtropical oceanic upwelling zones could have contributed significantly to the Pliocene warmth
During the early Pliocene, 5 to 3 million years ago, globally averaged temperatures were substantially higher than they are today, even though the external factors that determine climate were essentially the same. In the tropics, El Niño was continual (or "permanent") rather than intermittent. The appearance of northern continental glaciers, and of cold surface waters in oceanic upwelling zones in low latitudes (both coastal and equatorial), signaled the termination of those warm climate conditions and the end of permanent El Niño. This led to the amplification of obliquity (but not precession) cycles in equatorial sea surface temperatures and in global ice volume, with the former leading the latter by several thousand years. A possible explanation is that the gradual shoaling of the oceanic thermocline reached a threshold around 3 million years ago, when the winds started bringing cold waters to the surface in low latitudes. This introduced feedbacks involving ocean-atmosphere interactions that, along with ice-albedo feedbacks, amplified obliquity cycles. A future melting of glaciers, changes in the hydrological cycle, and a deepening of the thermocline could restore the warm conditions of the early Pliocene.
The current generation of coupled climate models run at the Geophysical Fluid Dynamics Laboratory (GFDL) as part of the Climate Change Science Program contains ocean components that differ in almost every respect from those contained in previous generations of GFDL climate models. This paper summarizes the new physical features of the models and examines the simulations that they produce. Of the two new coupled climate model versions 2.1 (CM2.1) and 2.0 (CM2.0), the CM2.1 model represents a major improvement over CM2.0 in most of the major oceanic features examined, with strikingly lower drifts in hydrographic fields such as temperature and salinity, more realistic ventilation of the deep ocean, and currents that are closer to their observed values. Regional analysis of the differences between the models highlights the importance of wind stress in determining the circulation, particularly in the Southern Ocean. At present, major errors in both models are associated with Northern Hemisphere Mode Waters and outflows from overflows, particularly the Mediterranean Sea and Red Sea.
Can atmospheric forcing of the ocean in high latitudes induce decadal variability in low latitudes? Most theoretical studies that have considered this question assign a critical role to adiabatic, advective, subsurface oceanic links between the tropics and extra-tropics. Observational evidence of such links is proving elusive. This study posits that given the constraint of a balanced heat budget for the ocean in a state of equilibrium, atmospheric forcing over a broad spectrum of frequencies in high latitudes can force decadal variability in low latitudes without any explicit evidence of oceanic links. The oceanic response to an abrupt change in diabatic forcing, a sudden increase in heat loss in high latitudes say, is characterized by two time-scales. The one, tw, is relatively short and is associated with planetary and coastal waves that propagate from the disturbed region to the equator (and then back to higher latitudes.) The other, td, is on the order of a few years and depends on diabatic processes responsible for increasing the oceanic heat gain in low latitudes. Through these processes the system is driven towards a new balanced heat budget in which the heat gain, mainly in the equatorial upwelling zones, equals the heat loss in high latitudes. When the forcing, rather than abrupt, is sinusoidal with period P, then the amplitude of the response depends on the ratio P/td. The response is modest when that ratio is small because the period P is too short for the ocean to adjust. As P gets larger compared to td, the amplitude increases, but explicit evidence of the waves that connect high and low latitudes is very hard to detect. The ocean acts as a low pass filter to the forcing with characteristic timescale td.
The salient feature of the oceanic thermal structure is a remarkably shallow thermocline, especially in the Tropics and subtropics. What factors determine its depth? Theories for the deep thermohaline circulation provide an answer that depends on oceanic diffusivity, but they deny the surface winds an explicit role. Theories for the shallow ventilated thermocline take into account the influence of the wind explicitly, but only if the thermal structure in the absence of any winds, the thermal structure along the eastern boundary, is given. To complete and marry the existing theories for the oceanic thermal structure, this paper invokes the constraint of a balanced heat budget for the ocean. The oceanic heat gain occurs primarily in the upwelling zones of the Tropics and subtropics and depends strongly on oceanic conditions, specifically the depth of the thermocline. The heat gain is large when the thermocline is shallow but is small when the thermocline is deep. The constraint of a balanced heat budget therefore implies that an increase in heat loss in high latitudes can result in a shoaling of the tropical thermocline; a decrease in heat loss can cause a deepening of the thermocline. Calculations with an idealized general circulation model of the ocean confirm these inferences. Arguments based on a balanced heat budget yield an expression for the depth of the thermocline in terms of parameters such as the imposed surface winds, the surface temperature gradient, and the oceanic diffusivity. These arguments in effect bridge the theories for the ventilated thermocline and the thermohaline circulation so that previous scaling arguments are recovered as special cases of a general result.
Studies of the effect of a freshening of the surface waters in high latitudes on the oceanic circulation have thus far focused almost entirely on the deep thermohaline circulation and its poleward heat transport. Here it is demonstrated, by means of an idealized general circulation model, that a similar freshening can also affect the shallow, wind-driven circulation of the ventilated thermocline and its heat transport from regions of gain (mainly in the upwelling zones of low latitudes) to regions of loss in higher latitudes. A freshening that decreases the surface density gradient between low and high latitudes reduces this poleward heat transport, thus forcing the ocean to gain less heat in order to maintain a balanced heat budget. The result is a deepening of the equatorial thermocline. (The deeper the thermocline in equatorial upwelling zones is, the less heat the ocean gains.) For a sufficiently strong freshwater forcing, the poleward heat transport all but vanishes, and permanently warm conditions prevail in the Tropics. The approach to warm oceanic conditions is shown to introduce a bifurcation mechanism for the north–south asymmetry of the thermal and salinity structure of the upper ocean.
This paper details a free surface method using an explicit time stepping scheme for use in z-coordinate ocean models. One key property that makes the method especially suitable for climate simulations is its very stable numerical time stepping scheme, which allows for the use of a long density time step, as commonly employed with coarse-resolution rigid-lid models. Additionally, the effects of the undulating free surface height are directly incorporated into the baroclinic momentum and tracer equations. The novel issues related to local and global tracer conservation when allowing for the top cell to undulate are the focus of this work. The method presented here is quasi-conservative locally and globally of tracer when the baroclinic and tracer time steps are equal. Important issues relevant for using this method in regional as well as large-scale climate models are discussed and illustrated, and examples of scaling achieved on parallel computers provided.
This paper discusses spurious diapycnal mixing associated with the transport of density in a z-coordinate ocean model. A general method, based on the work of Winters and collaborators, is employed for empirically diagnosing an effective diapycnal diffusivity corresponding to any numerical transport process. This method is then used to quantify the spurious mixing engendered by various numerical representations of advection. Both coarse and fine resolution examples are provided that illustrate the importance of adequately resolving the admitted scales of motion in order to maintain a small amount of mixing consistent with that measured within the ocean's pycnocline. Such resolution depends on details of the advection scheme, momentum and tracer dissipation, and grid resolution. Vertical transport processes, such as convective adjustment, act as yet another means to increase the spurious mixing introduced by dispersive errors from numerical advective fluxes.
Pacanowski, Ronald C., and Stephen M Griffies, 1999: The MOM3 Manual, GFDL Ocean Group Technical Report No. 4, Princeton, NJ: NOAA/Geophysical Fluid Dynamics Laboratory, 680 pp.
Griffies, Stephen M., Anand Gnanadesikan, Ronald C Pacanowski, V D Larichev, J K Dukowicz, and R D Smith, 1998: Isoneutral diffusion in a z-coordinate ocean model. Journal of Physical Oceanography, 28(5), 805-830. Abstract PDF
This paper considers the requirements that must be satisfied in order to provide a stable and physically based isoneutral tracer diffusion scheme in a z-coordinate ocean model. Two properties are emphasized: 1) downgradient orientation of the diffusive fluxes along the neutral directions and 2) zero isoneutral diffusive flux of locally referenced potential density. It is shown that the Cox diffusion scheme does not respect either of these properties, which provides an explanation for the necessity to add a nontrivial background horizontal diffusion to that scheme. A new isoneutral diffusion scheme is proposed that aims to satisfy the stated properties and is found to require no horizontal background diffusion.
Pacanowski, Ronald C., and Anand Gnanadesikan, 1998: Transient response in a Z-level ocean model that resolves topography with partial cells. Monthly Weather Review, 126(12), 3248-3270. Abstract PDF
Ocean simulations are in part determined by topographic waves with speeds and spatial scales dependent on bottom slope. By their very nature, discrete z-level ocean models have problems accurately representing bottom topography when slopes are less than the grid cell aspect ratio delta z/delta x. In such regions, the dispersion relation for topographic waves is inaccurate. However, bottom topography can be accurately represented in discrete z-level models by allowing bottom-most grid cells to be partially filled with land. Consequently, gently sloping bottom topography is resolved on the scale of horizontal grid resolution and the dispersion relation for topographic waves is accurately approximated. In contrast to the standard approach using full cells, partial cells imply that all grid points within a vertical level are not necessarily at the same depth and problems arise with pressure gradient errors and the spurious diapycnal diffusion. However, both problems have been effectively dealt with. Differences in flow fields between simulations with full cells and partial cells can be significant, and simulations with partial cells are more robust than with full cells. Partial cells provide a superior representation of topographic waves when compared to the standard method employing full cells.
Gnanadesikan, Anand, and Ronald C Pacanowski, 1997: Improved representation of flow around topography in the GFDL modular ocean model MOM 2. International WOCE Newsletter, 27, 23-25.
Li, X, P Chang, and Ronald C Pacanowski, 1996: A wave-induced stirring mechanism in the mid-depth equatorial ocean. Journal of Marine Research, 54(3), 487-520. Abstract
A wave-induced stirring and transport mechanism for the mid-depth equatorial ocean is proposed and examined using both analytic linear equatorial wave solutions and a fully nonlinear reduced-gravity model. The study of kinematic stirring using the linear solutions suggests that a superimposition of a few simple equatorial waves can lead to strong Lagrangian stirring and transport along the equator. In particular, a combination of an annual long Rossby wave and a high-frequency Yanai wave appears to be most effective in producing strong stirring in the interior equatorial region. Further investigations of stirring properties using an inverted, fully nonlinear reduced gravity shallow-water model support the results of the kinematic stirring study. By evaluating the finite-time estimates of Lyapunov exponents, we identified two regions where chaotic stirring is most active. One is the western boundary region where short Rossby waves likely play a dominant role in producing the strong chaotic stirring. The other is the equatorial waveguide where a low-frequency Rossby wave prescribes the pattern of the stirring geometry, and a high-frequency Yanai wave plays a role of stirring the fluid. The proposed stirring mechanism provides a plausible explanation of the observed chlorofluorocarbon distribution found in the mid-depth equatorial Atlantic Ocean.
Although the distribution of sunshine is symmetrical about the equator, the earth's climate is not. Climatic asymmetries are prominent in the eastern tropical Pacific and Atlantic Oceans where the regions of maximum sea surface temperature, convective cloud cover, and rainfall are north of the equator. This is the result of two sets of factors: interactions between the ocean and atmosphere that are capable of converting symmetry into asymmetry, and the geometries of the continents that determine in which longitudes the interactions are effective and in which hemisphere the warmest waters and the intertropical convergence zone are located. The ocean-atmosphere interactions are most effective where the thermocline is shallow because the winds can readily affect sea surface temperatures in such regions. The thermocline happens to shoal in the eastern equatorial Pacific and Atlantic, but not in the eastern Indian Ocean, because easterly trade winds prevail over the tropical Atlantic and Pacific whereas monsoons, with a far larger meridional component, are dominant over the Indian Ocean. That is how the global distribution of the continents, by determining the large-scale wind patterns, causes climatic asymmetries to be prominent in some bands of longitude but not others. The explanation for asymmetries that favor the Northern rather than Southern Hemisphere with the warmest waters and the ITCZ involves the details of the local coastal geometries: the bulge of western Africa to the north of the Gulf of Guinea and the slope of the western coast of the Americas relative to meridians. Low-level stratus clouds over cold waters are crucial to the maintenance of the asymmetries.
Wacongne, S, and Ronald C Pacanowski, 1996: Seasonal heat transport in a primitive equations model of the tropical Indian Ocean. Journal of Physical Oceanography, 26(12), 2666-2699. Abstract PDF
This work analyzes seasonal heat transport in an ocean-only numerical simulation of the Indian Ocean forced by realistic seasonal winds and surface heat fluxes north of 15°S, assuming no Indonesian Throughflow. The seasonal changes in the model circulation and temperature structure are found to be overall consistent with observations, despite flaws in sea surface temperature and mixed layer depth. The simulation confirms that the reversal of the monsoons and of the associated Ekman transports plays an important role in reversing the sign of the ocean heat transport seasonally causing, in particular, the Arabian Sea's drastic annual cooling, but it suggests that, south of 10°N, deep boundary currents must reverse as well. Most of the model heat transport is carried by a deep downwelling cell during the northeast monsoon and by a shallower upwelling cell during the southwest monsoon. An analysis of the three-dimensional circulation reveals that, in boreal summer, the net -1.2 pW (1 pW = 1015 W) cross-equatorial model heat transport derives mostly from a 20 x 106 m3 s-1 northward boundary current at intermediate levels (12.5°C) returned over the interior at the surface (27.5°C). In boreal winter, the net + 1 pW heat transport derives mostly from 10 x 106 m3 s-1 northward interior surface flow (27.5°C) returned in several deep southward boundary currents (5°C). It is argued that the +1 pW February heat transport value is realistic and that a deep overturning cell must therefore exist, otherwise the return branch of the relatively small February Ekman transport would have to occur at a negative transport-averaged temperature. Moreover, deep downwelling during the northeast monsoon occurs in the model because of a pattern of flow convergence at intermediate levels of the Somali Current that is consistent with direct observations. An approach toward assessing the location and the role of diabatic processes (which could be responsible for too deep a penetration of the downwelling cell) is tested, and a formal decomposition of the seasonal heat transport into diabatic and adiabatic components is suggested. Representing as a function of latitude and potential temperature an equivalent streamfunction associated with diffusion appears a promising step toward quantifying such diabatic heat transports on a seasonal basis.
Pinardi, N, Anthony Rosati, and Ronald C Pacanowski, 1995: The sea surface pressure formulation of rigid lid models. Implications for altimetric data assimilation studies. Journal of Marine Systems, 6, 109-119. Abstract PDF
The sea surface pressure formulation of the rigid lid primitive equation oceanic problem is reviewed and clarified. The geostrophic limit for the sea surface pressure equation is then considered and a new diagnostic relationship is found that relates the surface pressure to the barotropic and baroclinic components of the subsurface flow field. We demonstrate that a direct insertion in the model equations of sea surface information, such as that provided by satellite altimetry, does not produce changes in the subsurface dynamics due to the divergenceless nature of the barotropic flow field.
The geostrophic limit of the sea surface pressure field computed from a standard general circulation model of the world ocean is presented and the barotropic/baroclinic components of the asolute dynamic topography of the global general circulation are discussed.
Liu, Zhengyu, S G H Philander, and Ronald C Pacanowski, 1994: A GCM study of tropical-subtropical upper-ocean water exchange. Journal of Physical Oceanography, 24(12), 2606-2623. Abstract PDF
Experiments with an oceanic general circulation model indicate that the tropical and subtropical oceanic circulations are linked in three ways. Far from coasts in the oceanic interior, equatorial surface waters flow poleward to the southern part of the subtropical gyre, and then are subducted and returned in the thermocline to the upper part of the core of the Equatorial Undercurrent. Experiments with an oceanic general circulation model indicate that the tropical and subtropical oceanic circulations are linked in three ways. Far from coasts in the oceanic interior, equatorial surface waters flow poleward to the southern part of the subtropical gyre, and then are subducted and returned in the thermocline to the upper part of the core of the Equatorial Undercurrent. There is, in addition, a surface western boundary current that carries waters from the equatorial region to the northern part of the subtropical gyre. After subduction, that water reaches the equator by means of a subsurface western boundary current and provides a substantial part (2/3 approximately) of the initial transport of the Equatorial Undercurrent. The eastward flow in the Equatorial Undercurrent is part of an intense equatorial cell in which water rises to the surface at the equator, drifts westward and poleward, then sinks near 3° latitude to flow equatorward where it rejoins the undercurrent.
A global atmospheric general circulation model (GCM) coupled to an oceanic GCM that is dynamically active only in the tropical Pacific simulates variability over a broad spectrum of frequencies even though the forcing, the annual mean incoming solar radiation, is steady. Of special interest is the simulation of a realistically irregular Southern Oscillation between warm El Niño and cold La Niña states. Its time scale is on the order of 5 years. The spatial structure is strikingly different in the eastern and western halves of the ocean basin. Sea surface temperature changes have their largest amplitude in the central and eastern tropical Pacific, but the low-frequency zonal wind fluctuations are displaced westward and are large over the western half of the basin. These zonal wind anomalies are essentially confined to the band of latitudes 10°N to 10°S so that they form a jet and have considerable latitudinal shear. During El Niño the associated curl contributes to a pair of pronounced minima in thermocline depth, symmetrically about the equator in the west, near 8°N and 8°S. In the east, where the low-frequency wind forcing is at a minimum, the deepening of the thermocline in response to the winds in the west have a very different shape-an approximate Gaussian shape centered on the equator.
The low-frequency sea surface temperature and zonal wind anomalies wax and wane practically in place and in phase without significant zonal phase propagation. Thermocline depth variations have phase propagation; it is eastward at a speed near 15 cm s-1 along the equator in the western half of the basin and is westward off the equator. This phase propagation, a property of the oceanic response to the quasi-periodic winds that force currents and excite a host of waves with periods near 5 years, indicates that the ocean is not in equilibrium with the forcing. In other words, the ocean-atmosphere interactions that cause El Niño to develop at a certain time are countered and, in due course, reversed by the delayed response of the ocean to earlier winds. This "delayed oscillator" mechanism that sustains interannual oscillations in the model differs in its details from that previously discussed by Schopf and Suarez and others. The latter investigators invoke an explicit role for Kelvin and Rossby waves. These waves cannot be identified in the low-frequency fluctuations of this model, but they are energetic at relatively short periods and are of vital importance to a quasi-resonant oceanic mode with a period near 7 months that is excited in the model. The similarities and differences between the results of this simulation and those with other models, especially the one described in a companion paper, are discussed.
Two different coupled ocean-atmosphere models simulate irregular interannual fluctuations that in many respects resemble El Niño Southern Oscillation phenomena. For example, the spatial structure of various fields at the peaks of the warm El Niño and cold La Niña phases of the oscillation are realistic. This success indicates that the models capture certain aspects of the interactions between the ocean and atmosphere that cause the Southern Oscillation. The principal difference between the models, namely the prominence of oceanic Kelvin waves in one but not the other, causes the two models to differ significantly in the way El Niño episodes evolve, and in the mechanisms that cause a turnabout from El Niño to La Niña and vice versa. It is possible that the different processes that determine the properties of the simulated oscillations all play a role in reality, at different times and in different regions. Each of the models captures some aspects of what is possible. However, reality is far more complex than any model developed thus far and additional processes not yet included are also likely to have a significant influence on the observed Southern Oscillation.
Pacanowski, Ronald C., 1987: Effect of equatorial currents on surface stress. Journal of Physical Oceanography, 17(6), 833-838. PDF
A general circulation model of the tropical Pacific Ocean, which realistically simulates El Niño of 1982-83, has been used to determine how different initial conditions affect the model. Given arbitrary initial conditions (not in equilibrium with the wind) the model takes almost a year to return to a state in which the currents and density gradients are in equilibrium with the winds. Errors in the absolute value of the temperature persist far longer, however, indicating that accurate density data are essential initial conditions. If the correct density field is specified initially, but no information is provided about the currents, then the model recovers the currents within an inertial period, except for the eastern equatorial region. That region is affected by equatorial Kelvin waves which are excited because the model is initially in an unbalanced state. The currents associated with these waves are relatively modest and do not affect the density field significantly. Because of the large zonal scale of the thermal field in the tropical Pacific, three or four high resolution meridional density sections appear adequate for the initialization of the model. This result, however, takes into account neither the energetic waves, with a scale of 1000 km, that are associated with instabilities of the equatorial currents nor other high frequency fluctuations in the ocean.
Nonlinearities have a large effect on the circulation of the tropical Atlantic Ocean within a few hundred kilometers of the equator, both in the surface layers and at depth. Qualitative features of a nonlinear model that are absent from a linear model include energetic unstable waves in the western equatorial Atlantic and a westward surface jet that penetrates to considerable depths between the equator and 3°N. The largest quantitative difference between the nonlinear and linear models is the intensity of the westward surface flow at the equator. In a linear model it can be twice as fast as in a nonlinear model. Motion below the equatorial themocline, though sufficiently slow to be linear, is related to the surface forcing in a nonlinear manner because it is forced to a large extent by vertical movements of the thermocline. (Linear models assume that disturbances reach the deep ocean by propagating through a fixed thermocline.) In addition to the equatorial zone, nonlinearities affect the coastal zone of Africa south of the equator, where alongshore currents and zonal pressure gradients in linear and nonlinear models are different.
In a general circulation model of the tropical Atlantic Ocean, the northwestward flowing Brazilian Coastal Current is fed by the westward South Equatorial Countercurrent and in turn loses water to the eastward Equatorial Undercurrent and the eastward North Equatorial Countercurrent. The transport of the countercurrent decreases in a downstream direction primarily because of downwelling and then equatorward flow, in the thermocline, into the undercurrent. Some of the countercurrent water penetrates into the Gulf of Guinea, where it flows into the southern hemisphere. The transport of the Equatorial Undercurrent decreases because upwelling, which is most intense in the western side of the basin, transfers fluid into the surface layers to sustain divergent Ekman drift which is swept westward by the South Equatorial Undercurrent. The model has northward heat transport across all latitudes in the tropics. Seasonal variations in the transport are modest to the south of 5 degrees S and to the north of 15 degrees N. Across 8 degrees N, however, the transport varies from 1.5 x 1015 W in January and February to -0.1 x 1015 W in August. This result implies that the zonal bands 5 degrees S to 8 degrees N and 8 degrees N to 15 degrees N act as capacitors that are out of phase. In July, August, and September the heat gained from the southern hemisphere is stored in the 5 degrees S to 8 degrees N band where the thermocline deepens. During this period (when the Brazilian Coastal Current turns offshore near 5 degrees N) the thermocline between 8 degrees N to 15 degrees N rises as heat is lost across 15 degrees N. When the Brazilian Coastal Current flows continuously along the coast into the Gulf of Mexico, from December into May, it transports heat from the band 5 degrees S to 8 degrees N to replenish the heat stored between 8 degrees N and 15 degrees N and to sustain the heat flux across 15 degrees N.
In general circulation models of the seasonal cycle, westward propagating waves, with an approximate wavelength of 1000 km and period of 3 to 4 weeks, in the western equatorial Atlantic and eastern equatorial Pacific derive their energy from the kinetic and potential energy of the mean flow. There is intense downwelling the cold crests of the wave and upwelling in the warm troughs. The local meridional heat flux associated with the waves is of the order of 100 W m-2, but their contribution to the net heat transport across the equator is small. The waves are highly nonstationary in time and inhomogenous in space.
Philander, S G., and Ronald C Pacanowski, 1986: A model of the seasonal cycle in the tropical Atlantic Ocean. Journal of Geophysical Research, 91(C12), 14,192-14,206. Abstract
In the western tropical Atlantic, seasonal variations in the surface winds and in the ocean are dominated by an annual harmonic. A simulation with a general circulation model indicated that the response in the western side of the basin is an equilibrium one practically in phase with the local winds. It includes the following: large vertical excursions of the thermocline that have a 180 degree change in phase across 8 degrees N approximately; a change in the direction of the North Brazilian Coastal Current, which flows continuously along the coast between December and May but which veers offshore near 5 degrees N to feed the North Equatorial Countercurrent during the other months; and a seasonal reversal of the countercurrent. To the east of 30 degrees W, seasonal changes in the model have a prominent semiannual harmonic in phase with the local winds but only partially attributable to forcing at that frequency. The transients excited by the abrupt intensification of the southeast trade winds in May happen to have a phase essentially the same as that of the semiannual forcing. These transients decay by the end of the calendar year, so that the seasonal cycle that starts with the intensification of the winds in May can be treated as an initial value problem as far as the upper ocean, above the thermocline, is concerned. The winds along the equator determine the response of the surface equatorial layer in the Gulf of Guinea but play a minor role in the seasonal upwelling along the coast near 5 degrees N. That upwelling is strongly influenced by changes in both components of the wind, and in the curl of the wind, over the Gulf of Guinea.
Simulation of the seasonal cycle in the tropical Atlantic Ocean with a multi- level primitive equation numerical model yields remarkably realistic results including the separation of the Brazilian Current from the coast and the reversal of the Countercurrent.
During El Niño Southern Oscillation events modest anomalies amplify spatially and temporally until the entire tropical Pacific Ocean and the global atmospheric circulation are affected. Unstable interactions between the ocean and atmosphere could cause this amplification when the release of latent heat by the ocean affects the atmosphere in such a manner that the altered surface winds induce the further release of latent heat. Coupled shallow water models are used to simulate this instability which is modulated by the seasonal movements of the atmospheric convergence zones.
Measurements indicate that mixing processes are intense in the surface layers of the ocean but weak below the thermocline, except for the region below the core of the Equatorial Undercurrent where vertical temperature gradients are small and the shear is large. Parameterization of these mixing processes by means of coefficients of eddy mixing that are Richardson-number dependent, leads to realistic simulations of the response of the equatorial oceans to different windstress patterns. In the case of eastward winds results agree well with measurements in the Indian Ocean. In the case of westward winds it is of paramount importance that the nonzero heat flux into the ocean be taken into account. This heat flux stabilizes the upper layers and reduces the intensity of the mixing, especially in the east. With an appropriate surface boundary condition, the results are relatively insensitive to values assigned to constants in the parameterization formula.
Philander, S G., and Ronald C Pacanowski, 1981: The oceanic response to cross-equatorial winds (with application to coastal upwelling in low latitudes). Tellus, 33, 201-210. Abstract
Sea surface temperature variations observed in the eastern equatorial Atlantic and Pacific Oceans on seasonal, and possibly interannual (El Niño) time scales, may to a large extent be due to the variability of the local meridional winds. In a numerical model of the ocean, southerly winds cause low sea surface temperatures in the southeastern part of the basin because the coastal upwelling zone is extended far westward by (1) advection and (2) Rossby wave propagation which is important on time scales greater than a month. North of the equator sea surface temperatures are high. The thermocline has a trough near 3 degrees N where there is an intense eastward jet. A relaxation of the southerly wind causes a warming in the southeastern part of the basin primarily because of a zonal redistribution of heat by the South Equatorial Current and Countercurrent.
Oscillating wind with a period P induce variability with the following characteristics in the upper few hundred meters of the equatorial zone (5 degrees N to 5 degrees S) of the ocean. (1) P < 10 days: these winds fluctuate too rapidly to generate strong currents and excite primarily waves. (2) 10 days < P < 50 days: At these periods the winds generate intense equatorial jets in the upper 50 m, but at greater depths the variability has a small amplitude. Nonlinear eastward jets are more intense, are narrower, and are deeper than the corresponding westward jets so that winds with a zero mean value give rise to a mean eastward surface current. If the wind is always westward, then its fluctuating component intensifies the eastward equatorial undercurrent maintained by the mean winds. The surface flow is eastward and convergent when winds that are always westward go through a weak phase. (3) 50 days < P < 150 days: An eastward presure force exists sufficiently long to generate an intense eastward equatorial undercurrent. Variability has a large amplitude in the surface layers and in the thermocline. Eastward phase propagation associated with Kelvin waves is prominent in the upper ocean because the nonlinear currents impede the Rossby waves. (4) P > 150 days: The amplitude of variability is almost independent of frequency. An equilibrium response which is in phase with the forcing and which corresponds to a succession of steady states is approached asymptotically. These time scales are for a basin 5000 km wide. If the width of the basin exceeds 5000 km, then the 150 daytime scale increases. In the deep ocean below the thermocline, motion corresponds to propagating waves generated by the divergence of the nonlinear currents in the upper ocean.
Philander, S G., and Ronald C Pacanowski, 1981: Variability of SST in eastern equatorial oceans. Tropical Ocean-Atmosphere Newsletter, 6, 1, 7.
In response to the sudden onset of zonal winds the surface layers of the ocean accelerate in the direction of the wind. Motion is most intense near the equator where a jet forms within a week. The next stage in the evolution of equilibrium conditions is associated with wave fronts, excited initially at the coasts, that propagate across the ocean basin and establish zonal density gradients. Wave modes trapped in and above the strong shallow tropical thermocline because of internal reflection there are responsible for the adjustment of the upper ocean in low latitudes. These thermocline-trapped modes extend over a depth greater than that of the wind-driven surface currents and hence give rise to an undercurrent in the thermocline. This undercurrent is zonal and particularly intense near the equator, where it appears in the wake of an eastward traveling Kelvin or westward traveling Rossby wave after about 1 month. In the case of eastward winds, nonlinearities intensify the eastward equatorial surface jet and weaken the westward undercurrent. In the case of westward winds a different nonlinear mechanism intensidies the eastward Equatorial Undercurrent and weakens the westward surface flow. In a 5000-km winde basin, equilibrium equatorial currents are established about 150 days after the onset of the winds. The response time of the ocean below a depth of a few hundred meters is much longer. Winds with no spatial and a simple temporal structure generate currents with a complex vertical structure in the deep ocean. Closed current systems are possible in a confined forced region of an unbounded ocean; meridional coasts are not essential for their maintenance. The intensity of equatorial current is sensitive to dissipation parameters.
This paper is an analytical and numerical study of the response of the ocean to the fluctuating component of the wind stress as computed from twice-daily weather maps for the period 1973 to 1976. The results are described in terms of (time) mean and rms fields, frequency spectra and horizontal cross spectra, and local cross spectra between oceanic and atmospheric variables.
A forcing function with scales strictly larger than A forcing function with scales strictly larger than O(100 km) induces oceanic motion that is depth independent at periods between the inertial period and O(100 km) induces oceanic motion that is depth independent at periods between the inertial period and ~300 days. The dynamics is essentially linear so that rectified currents are small, the associated rectified transport amounts to at most 1-2 Sv in the western boundary layer. Root-mean-square currents are typically a few centimeters per second and are most intense in the western part of the basin, and near major topographic features. Fluctuations in the transport of the western boundary layer can be as large as 20-30 Sv. Three distinct frequency bands characterize the wind-induced barotropic fluctuations: 1) At periods between the inertial period and about one week the energy density increases steeply with decreasing frequency. Current spectra have a slope between -2 and -4. These forced waves can show an (imperfect) coherence between wind stress and the corresponding current components, and between atmospheric pressure and subsurface pressure. But spatial inhomogeneities in the wind field or bottom topography can destroy this coherence. 2) At periods between a week and a month planetary (or topographic) Rossby waves are dominant so that westward phase propagation is prominent. 3) At longer periods westward phase propagation is less evident and there is a time-dependent Sverdrup balance between meridional (cross-isobath) currents and wind stress curl. The spectra at these long periods are frequency independent (white) and the zonal (along-isobath) velocity component is more energetic than the meridional (cross-isobath) component.
Despite the high degree of idealization in the models, local coherence between oceanic and atmospheric variables is virtually nonexistent (except possibly at periods between 1 and 10 days) because of the wavelike structure of the oceanic response, the broadband stochastic character of the atmospheric variability, and inhomogeneities in the wind field and bottom topography.
It is proposed that fluctuations observed at site D north of the Gulf Stream are primarily atmospherically forced. At the MODE central mooring, however, there must be an additional energy source.
Anderson, D L., Kirk Bryan, A E Gill, and Ronald C Pacanowski, 1979: The transient response of the North Atlantic: Some model studies. Journal of Geophysical Research, 84(C8), 4795-4815. Abstract PDF
Four numerical experiments have been designed to clarify the role of stratification and topography on the transient response of the ocean to a change in wind forcing. The geometry and topography appropriate to the North Atlantic between the equator and 50 degrees N are used to make the study more appropriate to a real ocean. In all four experiments, zonally symmetric wind stresses are 'switched on' at the upper surface of a resting model ocean. Two short experiments, 1 and 2, with a duration of 100 days, are first discussed. These are for a homogeneous ocean with and without topography. The response in the flat-bottomed case can be described either in terms of planetary waves or basin modes, but when topography is present, no obvious wave propagation was identified. Higher-frequency basin modes are detectable, but their amplitude is much lower than that in the flat-bottomed case. They are damped out on a time scale of ~ 50 days. Two longer experiments, 3 and 4, are then analyzed. These are the analogs of 1 and 2, but stratification was included. The introduction of stratification for the ocean with topography leads to a new, longer time scale, not just for the baroclinic modes, but also for the barotropic. Despite the presence of topography, model analysis was found useful in analyzing the results. Propagation effects are analyzed, both on the moderately fast time scale of internal Kelvin waves and on the slow time scale of internal planetary waves. Kelvin waves are apparent along the equator, the northern boundary, and on the eastern coast in the Gulf of Guinea from the equator to 20 degrees N. They are not clearly visible anywhere on the west coast. Planetary waves can be detected in the interior both in the presence and absence of topography. When topography is present without stratification, the transport of the Gulf Stream is reduced from 30 to 14 million tons per second. This is a well-known result. With stratification there is no significant difference in transport between the case with or without topography.
A numerical experiment has been carried out with a joint model of the ocean and atmosphere. The 12-level model of the world ocean predicts the fields of horizontal velocity, temperature and salinity. It includes the effects of bottom topography, and a simplified model of polar pack ice. The numerical experiment allows the joint ocean-atmosphere model to seek an equilibrium over the equivalent of 270 years in the ocean time scale. The initial state of the ocean is uniform stratification and complete rest. Although the final temperature distribution is more zonal than it should be, the major western boundary currents and the equatorial undercurrent are successfully predicted. The calculated salinity field has the correct observed range, and correctly indicates that the Atlantic is saltier than the Pacific. It also predicts that the surface waters of the North Pacific are less saline than the surface waters of the South Pacific in accordance with observations. The pack ice model predicts heavy ice in the Arctic Ocean, and only very light pack ice along the periphery of the Antarctic Continent.
The poleward heat transport of the model is very sensitive to the strength of the circulation in the vertical-meridional plane. The heat transport is strongest in the trade wind belt where Ekman drift and thermohaline forces act together to cause a net flow of surface water toward the poles. At higher latitudes in the westerly belt the wind and thermohaline forces on the meridional circulation tend to oppose each other. As a result, the heat transport is weaker. Heat balance computations made from observed data consistently show that the maximum heat transport by ocean currents is shifted 10 degrees - A numerical experiment has been carried out with a joint model of the ocean and atmosphere. The 12-level model of the world ocean predicts the fields of horizontal velocity, temperature and salinity. It includes the effects of bottom topography, and a simplified model of polar pack ice. The numerical experiment allows the joint ocean-atmosphere model to seek an equilibrium over the equivalent of 270 years in the ocean time scale. The initial state of the ocean is uniform stratification and complete rest. Although the final temperature distribution is more zonal than it should be, the major western boundary currents and the equatorial undercurrent are successfully predicted. The calculated salinity field has the correct observed range, and correctly indicates that the Atlantic is saltier than the Pacific. It also predicts that the surface waters of the North Pacific are less saline than the surface waters of the South Pacific in accordance with observations. The pack ice model predicts heavy ice in the Arctic Ocean, and only very light pack ice along the periphery of the Antarctic Continent.
The poleward heat transport of the model is very sensitive to the strength of the circulation in the vertical-meridional plane. The heat transport is strongest in the trade wind belt where Ekman drift and thermohaline forces act together to cause a net flow of surface water toward the poles. At higher latitudes in the westerly belt the wind and thermohaline forces on the meridional circulation tend to oppose each other. As a result, the heat transport is weaker. Heat balance computations made from observed data consistently show that the maximum heat transport by ocean currents is shifted 10 degrees - 20 degrees equatorward relative to the maximum poleward heat transport by the atmosphere in middle latitudes. The effect of the zonal wind in enhancing poleward heat transport at low latitudes and suppressing it in middle latitudes is offered as an explanation.
