Upper Branch Routes of the Thermohaline Conveyor Belt
  Abstract >> Problem >> Strategy >> Routes >> Three Paths >> Warm, Cold and Tepid Routes >> Comparison with other GCMs >> A17 Comparison


 
 

Warm, Cold and Tepid Routes

The various routes can also be defined according to water masses present at the initial sections (i.e., DRAKE, ITFL and TAS):

WARM WATERS (i.e., surface waters) σ ≤ 26.5

COLD WATERS (i.e., intermediate waters) 27.0 < σ ≤ 27.6

TEPID* WATERS (i.e., mode waters) 26.5 < σ ≤ 27.

* The TEPID term has been chosen to contrast with the classical WARM and COLD denominations, referring to surface and intermediate waters, respectively.

 

 

WARM WATERS Waters flowing to EAQTL that are warm at the initial sections abound essentially at ITFL. Therefore, the classical identification of Warm Route with the Indonesian Passage holds for the ORCA-OPA results. This path represents 35% of the Upper Branch flow.

TEPID WATERS enter the Indo-Atlantic domain at all three initial sections, with almost equivalent amounts from DRAKE and TAS. This route accounts for 14% of the return flow.

COLD WATERS. Unlike classical hypotheses, the ORCA simulation shows that the Cold Water Route arises not only from DRAKE but also from TAS, even if the DRAKE contribution is the largest (5.6 Sv against 2.4 Sv). Nevertheless, it has to be noted here that we grouped in the Cold class also waters that are denser than 27.6. In fact, the ORCA simulation gives a contribution of 3 Sv of Intermediate waters and 2.6 of denser waters (i.e., s > 27.6; cf. figure below).

 

For DRAKE, the separation of trajectories according to water characteristics at the initial section permits to evidence the direct path to EQATL (i.e., without any detour in the Indian Ocean) undertaken by waters that are in the density range of mode waters at the Drake Passage.

Another interesting result suggested by the numerical simulation is the conversions undergone by the previously defined water masses as they enter the Indo-Atlantic domain. For example, waters that are intermediate at DRAKE are not necesseraly intermediate at EQATL. Therefore, the partitions in term of water masses change if they are defined at the origin section or at EQATL as it is shown in the two diagrams.

The ORCA simulation suggests that warm water at EQATL comes from both ITFL and DRAKE, in almost equal proportions (lower panel). This distribution is completely different from that based at the initial sections (upper panel) where ITFL appears as the absolute warm water provider. Hence, even though water entering the Indo-Atlantic basin through DRAKE is the coldest, at EQATL only 25% of Intermediate Water comes from DRAKE. A contribution that is equivalent to that from ITFL and half that from TAS. So, in the model, TAS appears as the major provider of Intermediate Water at EQATL, probably because TAS waters undergo less transformation processes than DRAKE waters.

 

Sabrina Speich