below ice is a lot warmer than above ice, so theoretically this helps in one dimension of the problem, which is probably the angle they're going for

practically it's still stupid because it's about as effective as moving a semi truck with a sewing needle

it's even stupider theoretically because the salt is gonna lower the melting point of the already-there ice, which likely has some freshwater snow precipitation on it

Ice isolates water from the cold air above though. And simply freezing lots of water makes the surface more reflective too, which mitigates further heating.

So as others have mentioned, the water underneath the ice is warmer than the air above it. That's a problem, because then the glacial sheets melt from the bottom up. If you pump enough water out from under the ice, not only will it refreeze in the much colder air above, it will eventually cause the glacier to sink until it makes contact with the ground again. That would remarkably slow the speed at which it melts, and thus help retain more ice longer which would prevent sea level rise and keep the reflective surface and reduce warming.

This sounds dumb, but the science is pretty cool. It would require a metric fuckton of pump capacity, though. This is likely a small proof of concept test to secure future funding.