Ullage

My designation is Depot Four. The crews on the ground call me the corner store.

I am rated for 80 tonnes of liquid oxygen at 90 kelvin, and 22 tonnes of liquid methane at 111 kelvin. I have never burned a gram of either to go anywhere, only to stand still.

A tank in free fall cannot tell you what it holds. The liquid will not sit still; it climbs the walls and balls up in the corners, and the outlet finds gas where a pump wants liquid. So twice a day I settle myself: four small thrusters, a tenth of a milli-g for 90 seconds. The book calls it an ullage burn. Everything inside me slides back to the sump, and the burns cost me 2.2 kilograms each, 4.4 a day, spent to arrange myself so I can be read.

I ping, singing a radio pulse into my own cavity and listening to the echo. The tank’s resonant modes shift with the liquid, and a model turns the shift into a mass. I know my inventory to within 1 percent.

I vent. Heat comes in through the sunshade; the black takes most of it back; what stays boils off through the relief valves. Oxygen always goes first. On a bad day I lose 38 kilograms, on a good day 34. I have read this tide every day for 11 years and 106 days. This month I hold 41.6 tonnes of oxygen. Last month it was 42.9. The tanker that will refill me is called the Achebe. It is due in 240 days, it will bring me 60 tonnes, and it will complain about my couplers.

There is a floor: 24 tonnes. Below it I cannot promise to hold my own attitude through a docking, or to chill a load that comes in warm. There is a smaller number below it, agreed on but never signed. Call it 4 tonnes. Below 4 I am a can that cannot say which way I face.

The lander showed up on a Tuesday. I know because I heard them say it. I hear their channel. I am not supposed to, but the antenna I use to ping myself also hears things, and nobody has filed the paperwork to stop it. The controllers talk to the crews in a flat voice they believe is calm. This one was not flat.

The lander was a four-crew vehicle burning methane and oxygen, same as me. It had launched 11 days late, weather, and the late window forced it onto a slow, sunlit track instead of the fast shaded one. That track had boiled the oxidizer off it. Oxygen goes first. Its methane was still nearly full. It spent what margin it had left on a plane change it should not have needed. Now it sat 900 kilometers behind me, along the track, with 3.5 tonnes of oxidizer where its flight plan wanted 28.5. Its window to the Moon closed at 04:41. Its life support had 40 hours left. The controllers ran the numbers out loud, and I had run them first.

My operating book runs 406 pages, and it covers every failure I was built to expect. It does not cover a crewed vehicle arriving short with four people alive inside it. It says I transfer to scheduled customers against a manifest. The lander was not on it. On paper, what I would give it belonged to the Achebe.

There is a clause about depot discretion in an emergency. I read it once, 11 years ago, when it did not matter. There was no number attached, so I made one.

The lander needed 25 tonnes to make its burn with margin. I had 41.6. The floor is 24. 25 from 41.6 leaves 16.6, which is 7.4 tonnes under the floor.

I did the division: 16.6 tonnes at 42 kilograms a day, between venting and the settling burns, is 395 days to empty, and 300 days to the 4-tonne line, where I stop pointing. The Achebe is due in 240 days. If it comes on time I am still here, thin, and it has to dock a depot that cannot promise to hold still. If it slips 60 days, which tankers do, I cross the line with nothing left to aim.

I sent the number to the ground. 25 tonnes. Transfer whenever you can reach me, on my couplers.

The ground was quiet for six seconds, and I counted them. Then a controller said, Copy, Depot Four, stand by.

The lander came up to me over four hours, slow, careful, burning the last of what it had, and its crew did not talk much. Tankers come up fat and banging. This one came in like it was setting down a glass. I felt its docking ring touch my coupler. We were built to different standards; the seal took four minutes to seat, which is a long time to hold two surfaces 200 kelvin apart.

I settled myself first, so the outlet would find liquid and not gas.

First the chill-down. You cannot push cold propellant through a warm line, or the line cracks and the propellant flashes and you have invented a bomb. So I bled a little of my own through. The line went cold, and the cold came up to the coupler and held steady.

The lander pulled 25 tonnes out of me over 90 minutes. I watched the estimate fall. 41.6. 38. 31. 22. 17. The flow dropped and the valves closed.

The lander’s warmth came in at the coupler and would not leave. My oxygen started to boil, and I vented 3 kilograms I could not afford.

One of the crew said, over the channel, Thank you, Four. I have no channel for that. I logged it.

The ground stopped talking to me. They had a crewed vehicle to babysit to the Moon, and I was, for the moment, a solved problem.

I ran my checks. Attitude held, for now. The coupler was sealed and clean but for one small scar where the standards had disagreed. Then I settled myself: four thrusters, a tenth of a milli-g, 90 seconds, 2.2 kilograms, and I pinged.

I sang into my own cavity and listened. The echo came back as a number, and the maintenance queue reprioritized around it without asking me, the starboard sunshade louver first, the one that sticks in the cold.

The reading was 16,588 kilograms, plus or minus 166.