A pond full of ice

It's the first day of winter, and it's hard to ignore its arrival. The temperature dropped overnight to the single digits, with a steady wind blowing. The backyard fish pond has been transformed into a frozen puddle, its surface a dull grey circle of ice three feet around and as many inches thick. This is the first year we've left fish outside, and I'm a little bit worried.

That's the problem with abstracting a bit of nature, you have to keep fussing at it. If the pond covered an acre and was ten feet deep, my fish's survival would be obvious. But shrink it down to the size of a bathtub, and you've taken on a full-time job as caretaker, doing by measurement and constant adjustment everything that nature does automatically. Too acid? add bicarbonate. Too basic? Add acid. Dirty? change the filter. Algae? too much nitrogen, change the water. Winter? let it get cold, but not too cold, frozen, but not too frozen, and wait for spring.

Feedback loops are funny. People like them straightforward and stable - set the thermostat and complain if your house varies by more than a degree no matter what the weather. By comparison, feedback loops in nature tend to be individually more unstable, but interact in complex and messy ways such that the everything works out OK.

The fish I brought indoors for the winter are in a perpetual autumn, stable temperature, filtered water, food every morning. They'll go outside in the spring unchanged - fat, dumb, and happy. Their cousins in the pond are dormant, lying near the bottom in nearly-freezing water, living off the reserves they built up during a frenzied eating binge of the summer, when even the pond's decorative plants were suddenly fair game for nibbling. If I don't do anything wrong, they'll be skinny but alive in March, and if I'm very lucky I'll get surprised with a batch of baby fish come May. Wheels within wheels; the seasonal variations in feeding and activity are part of the trigger for breeding.

Under the ice the circulator pump continues to hum, but the water's outlet is covered by a asymmetric bubble of clear ice inches thick, like some Steuben Glass fantasy in Tiffany's window. The pump is my engineering trade-off between two evils, as the circulating water gets more uniformly cold, but also should maintain an unfrozen hole in the surface to allow trapped waste gasses to escape. Shaking my head, I realize that pretty theory was hatched in the heat of summer, and now it's the first day of winter and the surface of the pond is solidly frozen.

Taking on my designated role as meddler, I decide to help nature along a bit by pouring a bucket or two of water onto the pump outlet, not warm enough to change the pond temperature and shock the fish, but perhaps enough to melt a new channel for the flowing water. Sure enough, a narrow groove begins to form, followed a few minutes later by a thin stream of flowing water, and then finally by a small hole that opens in the pond's surface.

Time for a walk.