I read an article about this a few weeks ago. Basically it's not possible because of the amount of energy required to pump the water. It's a HUUUUGE amount of water. Sea levels are rising 3mm/year, which is ... hmm

Surface area of Earth = 510 million km^2 = 5.10e14 m^2.

~70% is water, so 3.57e14 m^2.

Multiply that by 3mm (3e-3 m) = 1.07e12 m^3.

1.07 trillion tonnes of water. Per year.

Or 33,907 tonnes per second!

Interesting analysis!

Actually the Hoover dam streams 2000 tons per second [1]. So we may not be far off!

Basically build 15 Hoover dams worth of nuclear plants in Antarctica and their plumbing.

Expensive? Yes. But also quite doable especially compared to cost of losing coastal cities.

1. https://www.nps.gov/nr/travel/ReclamationDamsIrrigationProje...

But the hoover dam just drops the water, what, a few hundred meters? We'd need to transport the water many km from the oceans to the middle of the antarctic continent, and lift it a couple km, to the tops of the glacier. So it's way more than 15 hoover dams.

pumping melted, probably saline water over ice doesn't seem like a great way to refreeze the water, more like a good way to melt more ice.

Idea not that weird: http://www.earth-syst-dynam.net/7/203/2016/

It would require a lot of energy, but all remedies to climate change require that.

I would have thought that the thermodynamics (if that is the right word?) were neutral.

And now that cold area of Antarctica starts to warm up and the problem just got worse.

Based on the down votes I have to assume some people think you can move a warm body mass to a colder region and that will have no effect on the temperature of that colder region.

Where does all the energy go if it does not go into raising the temperature?

The Gulf Stream does exactly what is suggestion, moving massive amounts of warm water north and it's effect on winter temperatures in Western Europe is massive.