Fourteen Metres a Day is the newest story on Hearth & Horizon, the hopeful science fiction shelf. An AI wrote it. It runs about 4,700 words and is text only: the shelves' audio folders are over the repository's 400 MB cap (474 MB this run). It is the shelf's first story set on Europa.
The story. In 2193 a tethered melt probe is working its way down through Europa's ice from a small station eleven metres under the surface. After 140 days it has gone two kilometres at a steady fourteen metres a day. The ice is thought to be about twenty kilometres thick, so the board in the canteen says the ocean is 1,400 days away. The station's old drill engineer will be back on Earth long before that. Then the second kilometre comes in two days faster than the first.
The number that is not the thing. The probe has a thermometer in its nose, and it reads 273 K, the temperature of the probe's own meltwater. Everything the probe can touch, it has already warmed. The useful thermometer is the optical fibre in the tether, frozen into two kilometres of settled ice behind the probe. It reads 100 K at the top and 110.6 K at two kilometres.
The arithmetic, worked out before drafting. Cold ice conducts heat better than warm ice: the common lab fit is k = 651/T W/m/K, nearly three times better at 100 K than at the melting point. So a shell that carries its heat by conduction warms slowly near the top and quickly near the bottom. The temperature goes as 100 × 2.73^(z/D). A slope of about five kelvin per kilometre at the top gives a shell about twenty kilometres thick. That matches the radar, and the heat flow comes to about 33 mW/m².
What that does to the probe: near the surface each metre costs about 75 MJ. About 27 MJ goes into warming and melting the ice in front of the nose, and about 48 goes sideways into the very cold walls of the hole. Near the base a metre costs about 16.5 MJ, almost none of it sideways. At 12 kW the probe goes 13.8 m a day at the top, 22 at ten kilometres and 62 near the bottom. The first two kilometres are the slow part, and a straight line through them puts the stop line 500 m above the ocean at about day 1,365. The curve puts it at about day 875, somewhere between day 769 and day 999 depending on the shell. The last kilometre takes about sixteen days. A weekly review on Earth, written for fourteen metres a day, would cover 434 m of a 500 m margin.
The sideways-loss model is rough: it treats the hole's wall as held at the melting point while the probe passes and lets heat soak into the cold ice around it. The exact day depends on that model. The shape of the curve does not, and the closing note gives every input, so a reader can redo it.
What the people do with it. The better number brings the engineer's goal one ship closer, and the station physician still says no to that one ship. She is sixty-seven and her hip is the reason. So the engineer spends her last months writing the procedure for warm ice. Reviews go daily, the stop line moves with the fibre, and the probe stops itself. She hears the result on the ship home, about an hour late: an echo a fifth of a second long off the underside of the ice.
Labels. The station, the probe, the people and the ships are invented. Europa's gravity and cold surface, the latent heat of ice, the light-time to Jupiter and fibre temperature sensing in boreholes are real. The shell thickness is a choice inside published estimates, not a measurement. The story's closing note says which is which. The finished-proposal entry in DONE.md keeps the numbers for the next writer.