Surprises in Ice
Earth has done its seasonal change trick again, and we northerners are bracing for winter. It's time to think of ice.
Ice will soon glaze everything from puddles to lakes, but its commonplace nature doesn't mean it's uninteresting. The freezing process alone has a few surprises. Fresh water doesn't freeze precisely at zero Celsius or thirty-two degrees above zero Fahrenheit, because the transformation of water to ice is a change of phase: the liquid becomes a solid, and that takes extra energy. The transition to solid state requires the loss of 80 calories per gram of water---real calories, not dieter's calories, which are properly kilocalories and a thousand times larger.
The small calorie still represents an impressive amount of energy. It is defined as the amount of heat required to change the temperature of one gram of water by one degree C. (That's putting it roughly; in exact terms, it's the amount of heat needed to raise the temperature of one gram of water, under a pressure of one standard atmosphere, from 14.5 to 15.5oC. For all practical purposes, it's a case of add a calorie, the gram of water goes up one degree in temperature; subtract that calorie, its temperature goes down by one degree.) Thus, a body of fresh water at 0oC must lose as much heat energy to freeze into ice as it would lose cooling from a temperature of 80oC down to 0oC (176 to 32oF).
If water were like most liquids, it would become more and more dense as its temperature goes down. That's an effect easily seen in a jug of syrup. And if water were like most other substances, it would be heavier in its solid phase than in its liquid state. Consider that syrup again--the crystallized portion is heavier, so it settles to the bottom of the jug.
So, if water were not the odd substance that it is, a pond would continue to cool until all its water reached the freezing point; then it would solidify from the bottom up. But water only behaves like most liquids (that is, contracting and becoming more dense with decreasing temperature) down to a temperature of about 4oC (39.2oF). As its temperature decreases further, it expands. Finally, when it freezes, its volume jumps by about 10 percent. That's why ice cubes rise to the top of a glass of water--ice is less dense than water of any temperature.
This peculiar behavior dictates how a pond freezes. As water at the surface loses heat to the cold air above, it becomes denser. Because it is more dense, this colder water sinks toward the bottom. That motion displaces warmer water upward, where it cools and then sinks in turn.
This tidy cycle exchanging cold water for warm can continue only until all available water in our hypothetical pond chills to that magic temperature of 4oC. Then the water that happens to be on top is stuck there, because water cooled below that temperature is Iess dense. The surface water will float, a puddle atop a pond, on the 4o water. Before long, it freezes into a buoyant lid, sealing in the water below and slowing its rate of heat loss to the frigid air.
Ice isn't a particularly good insulator, so water in contact with the ice lid freezes to it, at a pace governed by the thickness of the ice and the temperature of the air above. The pond freezes from the top down. If it's deep enough, free water will remain at the bottom all winter---a very good thing for fish as long as they're comfortable at a temperature of 4 degrees Celsius.
