It only took 25 years, but my thesis topic is now becoming newsworthy! Gluttons for punishment can see at least the abstract at A model of the formation of high-salinity shelf water on polar continental shelves. Which is aimed at one of the important ingredients for AABW (Antarctic Bottom Water).
I've been reluctant to blog about the topic because it is, after all, my baby and I'm sorely tempted to post at excruciating length and detail. (Not that there aren't other people who have studied the topic before or since, but I'm one of the people who has.)
I'll take this note as opportunity to get in to some detail about the weirdness that is sea water, and come to the climate change, carbon dioxide burial, and heat burial, aspects later. The story of AABW turns on some odd facts about how sea water behaves in Antarctic conditions. Not least, it can go below freezing.
The freezing point of sea water, like any material, depends on the pressure it is under. At the surface, Antarctic waters typically have a freezing point of -1.84 C. Yet, if you drag this water underneath an ice shelf, which Antarctica has some impressive examples of, you'll get temperatures down to about -2.3 C. This is 'Ice Shelf Water' (ISW; we're really not very creative). The 0.4 C difference may seem small, but all that matters is whether the water gets denser than what is formerly below it. This is 0.4 C drop in freezing point for being at the bottom of an ice shelf something like 1000 meters (yards) thick. Being this far down in the ocean means the water is under a pressure of about 100 times what you are breathing (100 atmospheres). The water cools to the local freezing point as it melts the bottom of the ice shelf. This sort of melting is also turning out to be extremely important for the fate of the Antarctic ice sheet.
The easier way of making water denser in the Antarctic is to freeze some sea ice. The ice crystals mostly leave behind the ocean's salt. (Do the experiment, maybe also with some food coloring in the water.) The sea ice then floats, and the salt makes the sea water denser. The denser water sinks. Just how far depends on how dense it is, and how dense water lower down is. Again, we're looking at small changes. The so-called 'Winter water' is near freezing and about 34.5 parts per thousand salt. The High Salinity Shelf Water (HSSW) is what you get after freezing sea ice from Winter Water. It is also near freezing, but is about 34.7 parts per thousand salt. Whew, a whole 0.2 parts per thousand (200 parts per million) controlling the process!
To get AABW, you then mix ISW or HSSW with the waters that are farther away from the Antarctic ice shelves and coast. AABW isn't quite as dense as the ISW or HSSW, but this mixing means that there is a lot more AABW than ISW or HSSW (call them Andy, Ignatz, and Henry if you like, the acronyms don't matter, just that we're talking about 3 players in the climate system). The AABW then flows off to the Atlantic ocean (primarily) along the bottom. Because it is at the bottom, the North Atlantic Deep Water (NADW) flows above it -- and above the mid-ocean ridges, and then in to the rest of the deep and bottom of the ocean.
But this only happens because of a different strange fact about how sea water behaves, and because of history. Same as freezing point depends on pressure, so does density. This isn't too surprising -- if you squeeze a sponge, it gets more compact (denser). But ... add another character to our story, the very warm, very salty water that flows out from the Mediterranean sea in to the Atlantic. If we compute the density, at the surface, of this Mediterranean Sea water and the Antarctic Water, we see that it is the Mediterranean which is denser, so 'ought' to be at the bottom of the ocean.
So one last weirdness -- it is easier to compress cold water than warm water. At and below about 2000 meters (200 atmospheres pressure), it is the Antarctic water which the densest. The history is this: If the ocean were filled from bottom to 1800 meters depth with the warm salty Mediterranean sea water, Antarctic water would not be able to sink to the bottom. We'd have a very different ocean. Among many other differences: sea level would be higher, and much less carbon would be in the ocean.