I read that thermohaline circulation is caused by cold and saline water sinking, while warm and less dense water rises. link

I understand that saline water is more dense, as there is a higher mass due to dissolved salts.

However, decreasing the temperature of water decreases the solubility, so in theory, shouldn't cold water hold less salt than warm water?

Vice-versa, shouldn't warm equatorial water increase in density as it is able to hold more salt?


Warm water does hold more salt - at saturation, but seawater isn't even close to saturation. Surface sea water gets warm, starts to evaporate (hence clouds) and therefore gets denser. So it sinks to the ocean floor, where it cools down to about 4 deg Celsius, and gets denser still. But it still is a very long way from being saturated with salt, so 'how much salt the water can hold' doesn't come into it.

  • 1
    $\begingroup$ Surface sea water gets warm, starts to evaporate (hence clouds) and therefore gets denser. So it sinks to the ocean floor, Does this mean in equatorial water where there is a high amount of evaporation and salinity in surface waters, the surface equatorial waters sink down to the ocean floor at the equator? $\endgroup$ – G. Gip Jul 19 '16 at 8:54
  • $\begingroup$ @G.Gip: there is a lot more heating at the surface at the equator, and that largely cancels the effect you're talking about. Warmer water spreads out towards the poles (not directly, there are other forces involved too), where it cools and sinks. This is known as (Ant)arctic bottom water formation. See for example mail.tku.edu.tw/086138/EnvFutures/WebPages/Global%20warming/… $\endgroup$ – naught101 Jul 20 '16 at 1:50

Usual salinity and solubility of NaCl

An usual salinity in the world's major oceans is 35 g/kg. The solubility of sodium chloride in water is around 360 g/kg (Sodium Chloride, Solubility in water). Therefore, the solubility is not the limiting factor.

Sea salt consists mainly of sodium chloride (NaCl). Therefore, this approximation should be OK.

Mixing of salt and water

The salt in the sea water 'moves together' with the water. Thus, when warm water flows from the equator to the poles it transports salt to the poles. At the poles water cools down and sinks (together with salt).

If we had a direct relation between sea surface temperature and salinity, in some region we would need a net flux of salt in the opposite direction to the water net flux.

Global salinity gradients

Global salinity gradients exist. The salinity is above the global mean salinity in some regions -- particularly in warm regions --, whereas in other regions the salinity is below this mean. These salinity gradients are caused by a net loss of oceanic water (evaporation > precipitation + fresh water inflow; enhanced salinity) in warm regions and by a net input of water by (evaporation < precipitation + fresh water inflow; reduced salinity), respectively. These gradients are not related to solubility of salt.

  • 1
    $\begingroup$ Could you explain/clarify that final sentence? I was following your answer clearly until then. $\endgroup$ – G. Gip Jul 19 '16 at 8:52
  • $\begingroup$ I reformulated the final sentence. What I mean is that if there was higher salinity at the equator (conjunctive) then it would be necessary that the salt is transport somehow to the equator. Hence we would have a net flux of salt from the poles to the equator. It would be in inverse direction then the water flows. This won't work. $\endgroup$ – daniel.neumann Jul 19 '16 at 10:18
  • $\begingroup$ But the equator does have a higher salinity than the poles, no? A look at a annual salinity map of the ocean shows this. $\endgroup$ – G. Gip Jul 22 '16 at 17:27
  • 1
    $\begingroup$ @G.Gip you are right. I should have though about it before writing it :-) . Corrected and added a third section. $\endgroup$ – daniel.neumann Jul 27 '16 at 9:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.