I came across this article, which le me to this other one.
Very interesting stuff, I learnt a lot

Could you explain to us some of what you learned?

That's especially true when it comes to athletics.

crank wrote:That's especially true when it comes to athletics.

What do you mean?

Sports lad, sports, what do you think athletics means? You're in for quite a treat if you can come out of the [water] closet.

If water ice were more dense than liquid water, the pond would freeze from the bottom up, and everything in it would die during the winter. Now you should read up on 'snowball Earth'.

If water ice were more dense than liquid water, the pond would freeze from the bottom up

Only if it was cold enough to freeze the whole pond, since this doesn't happen now it wouldn't under those circumstances either. What would happen is that ice would start to form at the surface, then sink, and melt before it got to the bottom. This circulation may make a bit more heat transfer happen, but it wouldn't likely make the difference between ponds/lakes that only freeze a few inches deep now getting so much colder that ice crystals sinking as soon as they can break surface tension would reach the bottom intact and stay that way.

zulumoose wrote:

If water ice were more dense than liquid water, the pond would freeze from the bottom up

Only if it was cold enough to freeze the whole pond, since this doesn't happen now it wouldn't under those circumstances either. What would happen is that ice would start to form at the surface, then sink, and melt before it got to the bottom. This circulation may make a bit more heat transfer happen, but it wouldn't likely make the difference between ponds/lakes that only freeze a few inches deep now getting so much colder that ice crystals sinking as soon as they can break surface tension would reach the bottom intact and stay that way.

When the surface waters cool, why would the water at the surface remain cooler than water below it if convection were possible? We are trying to imagine a world where ice is denser than water, and in that case, let's suppose there is not a density maximum above the freezing point past which cool surface water could not convect creating downwelling and bottom currents that are above the freezing point.

Given that scenario, why do you think that water at depth would be warmer than surface water, if convection were possible. We know that if the pool is really shallow, it just freezes solid, like a puddle on the pavement. The reason why the "freezing from the bottom up" scenario is pictured is that the water would be density-stratified and the stratification would be monotonic in density: The densest water would always be the coldest, until (denser, colder) ice began to sink into it.

Floating ice insulates the water body. If ice sank there would be no insulation and nothing to stop the water body progressively freezing.

Fenrir wrote:Floating ice insulates the water body. If ice sank there would be no insulation and nothing to stop the water body progressively freezing.

Yes, that's a concise description of the process. I think zulumoose is wondering about the heat budget in the bottom water, and his hypothesis is that it would melt the ice once it sank. That neglects a couple of factors.

The heat variation is atmospheric, winter doesn't start underground.

drops below freezing will always start in the air above ground, and it will take a lot of time for that to transfer to the ground under the pond. In fact since white snow reflects the suns radiation a pool with "alternative" water that would freeze only from the bottom up may be warmer than an adjacent pool with water as we know it.

And since the hypothetical pools don't already freeze solid with normal water, it is already a fact that "water at depth would be warmer than surface water", since it demonstrably isn't frozen. There shouldn't be a dispute over this.

zulumoose wrote:The heat variation is atmospheric, winter doesn't start underground.

drops below freezing will always start in the air above ground, and it will take a lot of time for that to transfer to the ground under the pond. In fact since white snow reflects the suns radiation a pool with "alternative" water that would freeze only from the bottom up may be warmer than an adjacent pool with water as we know it.

And since the hypothetical pools don't already freeze solid with normal water, it is already a fact that "water at depth would be warmer than surface water", since it demonstrably isn't frozen. There shouldn't be a dispute over this.

That's indeed the case in the real world. I thought we were talking about water that doesn't have a density maximum above freezing. You do understand density stratification and convection, do you not?

Please forgive the metaphor of 'freezing from the bottom up'. Maybe it would make more sense to say that ice would pile up at the bottom. In other chemical systems without the density anomaly, stratified crystallization is indeed observed with the densest stuff (or rather the earliest crystal fraction) in the bottom layers, like layered igneous intrusions, crystals setting out of a liquid that is less dense than the crystals. Those are multiphase, multicomponent systems, and the best we have in water world is dissolved salt. I'm not an expert in thermohaline convection, but I have heard of it, and it's one of those things that makes oceanography so interesting.

The conjecture wasn't about whether water is denser or warmer at depth or precisely where water would freeze.

The conjecture was about what would happen if ice sank once frozen.

Ice sinking changes the entire dynamics of the system.

And that's without trying to think about other factors, like if ice sank then the density of water most likely wouldn't be highest at 4C like it is and the effects that would have.

zulumoose wrote:...And since the hypothetical pools don't already freeze solid with normal water, it is already a fact that "water at depth would be warmer than surface water", since it demonstrably isn't frozen. There shouldn't be a dispute over this.

Except that the freezing point of water depends on the concentrations of any impurities in it, including sea salt:
https://en.wikipedia.org/wiki/Freezing-point_depression

Freezing-point depression is the process in which adding a solute to a solvent decreases the freezing point of the solvent. Examples include salt in water, alcohol in water, ...

zulumoose wrote:...

And since the hypothetical pools don't already freeze solid with normal water, it is already a fact that "water at depth would be warmer than surface water", since it demonstrably isn't frozen. There shouldn't be a dispute over this.

Behold! Shallow pools of water frozen through - welcome to the world of ice cubes.

Surely, the most important peculiarity of water is that it enable life to exist in the first place, not that it protects it.

DavidMcC wrote:Surely, the most important peculiarity of water is that it enable life to exist in the first place, not that it protects it.

Surely.

Actually since water is so common, the only peculiar thing is that we find its behaviour peculiar at all, instead of wondering why all the less common substances don't act the same.

zulumoose wrote:Actually since water is so common, the only peculiar thing is that we find its behaviour peculiar at all, instead of wondering why all the less common substances don't act the same.

Water isn't really that abundant, mass over mass if you look at the entire planet or planetary system. We find its behavior peculiar because we understand something about its properties and something about the properties of other kinds of solvents. Nevertheless, over a range of temperature and pressure that is well-known, water is a liquid. There ya have it. You can titillate yourself with the sense of how peculiar your circumstances are, or you can learn some physical chemistry. You can even do both.

I'm still coming to terms with the peculiar fact that there exists a wiki on "How to remove ice cubes from a tray."