368
Equatorial Dynamics of the Thermocline: The Equatorial Undercurrent
strength of the undercurrent near the eastern boundary. Although the solution
requires the undercurrent to be depleted by the time it reaches the eastern
boundary, the brusqueness with which this takes place is remarkable and reflects the increasing influence of the entrainment on the shallower structure of
the current in the eastern part of the basin.
6.7 Numerical Models
The connection between the dynamics of the EUC and the subtropical thermocline is clearly demonstrated by the representation of the EUC as an inertial
boundary-layer extension to the equator of the solution for the midlatitude
thermocline. The midlatitude thermocline and the EUC are two overlapping
elements of a single, complete picture of the general circulation. Nevertheless,
the proposed boundary-layer solutions themselves are incomplete and do not
yield a completely closed circulation. Only plausibility arguments have been
used to link the equatorial flow with the western boundary layer, for example.
Similarly, the termination of the undercurrent by entrainment into the upper
mixed layer is as much a hypothesis as a definitive result of the theory since the
entrainment process is not adequately represented in the simple theories presented. One way to examine the consistency of the hypothesized dynamical
elements of the conceptual model is to construct a numerical model whose
dynamics is sufficiently complete to allow us to check at least the self-consistency of these individual elements. Of course, care must be taken that the
numerical model itself is adequate, and in particular it must be able to describe
both the equatorial and midlatitude circulations well enough to persuasively
examine their linkage.
The early pioneering model of Cane (1979) is especially significant in that it
carefully emphasized the importance of the zonal variation of the flow for the
dynamics. Its vertical structure, however, consisted of a layer of a single density. The layer is heuristically subdivided into an upper mixed layer communicating with a single laminar layer below. With the added limitation of the
model to a latitude band of 15° on either side of the equator, it is difficult to use
the model to examine the connection of the equatorial flow with the midlatitude thermocline. Nevertheless, the model is successful in generating an
equatorial boundary layer form for the EUC whose dynamics is largely inertial,
and whose velocity satisfies the scaling relation (6.3.12). It is an important
indication that the ideas of Fofonoff and Montgomery and the inertial theory
outlined in Section 6.4 are relevant to the problem in which the finite zonal
extent of the basin is explicitly considered.
A more recent numerical model constructed for the expressed purpose of
examining the link between the equator and the midlatitude regions has been
presented by McCreary and Lu (1994). The model consists of two moving
layers over a resting abyss. The model is an interesting hybrid model containing
Equatorial Dynamics of the Thermocline: The Equatorial Undercurrent
strength of the undercurrent near the eastern boundary. Although the solution
requires the undercurrent to be depleted by the time it reaches the eastern
boundary, the brusqueness with which this takes place is remarkable and reflects the increasing influence of the entrainment on the shallower structure of
the current in the eastern part of the basin.
6.7 Numerical Models
The connection between the dynamics of the EUC and the subtropical thermocline is clearly demonstrated by the representation of the EUC as an inertial
boundary-layer extension to the equator of the solution for the midlatitude
thermocline. The midlatitude thermocline and the EUC are two overlapping
elements of a single, complete picture of the general circulation. Nevertheless,
the proposed boundary-layer solutions themselves are incomplete and do not
yield a completely closed circulation. Only plausibility arguments have been
used to link the equatorial flow with the western boundary layer, for example.
Similarly, the termination of the undercurrent by entrainment into the upper
mixed layer is as much a hypothesis as a definitive result of the theory since the
entrainment process is not adequately represented in the simple theories presented. One way to examine the consistency of the hypothesized dynamical
elements of the conceptual model is to construct a numerical model whose
dynamics is sufficiently complete to allow us to check at least the self-consistency of these individual elements. Of course, care must be taken that the
numerical model itself is adequate, and in particular it must be able to describe
both the equatorial and midlatitude circulations well enough to persuasively
examine their linkage.
The early pioneering model of Cane (1979) is especially significant in that it
carefully emphasized the importance of the zonal variation of the flow for the
dynamics. Its vertical structure, however, consisted of a layer of a single density. The layer is heuristically subdivided into an upper mixed layer communicating with a single laminar layer below. With the added limitation of the
model to a latitude band of 15° on either side of the equator, it is difficult to use
the model to examine the connection of the equatorial flow with the midlatitude thermocline. Nevertheless, the model is successful in generating an
equatorial boundary layer form for the EUC whose dynamics is largely inertial,
and whose velocity satisfies the scaling relation (6.3.12). It is an important
indication that the ideas of Fofonoff and Montgomery and the inertial theory
outlined in Section 6.4 are relevant to the problem in which the finite zonal
extent of the basin is explicitly considered.
A more recent numerical model constructed for the expressed purpose of
examining the link between the equator and the midlatitude regions has been
presented by McCreary and Lu (1994). The model consists of two moving
layers over a resting abyss. The model is an interesting hybrid model containing