How do you simulate tidal currents in an O'Neill Cylinder to maintain a healthy biome for aquatic life?

The O'Neill Cylinder contains artificial rivers, freshwater lakes, and "Oceans" which are just 60m deep large bodies of saltwater simulating continental shelf biomes.

This question has a lot of possible answers so to narrow it down I am looking for a solution that addresses the following in order of importance:

  1. Consumes the least possible energy and resources to maintain.
  2. Can be built in as many different kinds of places as as possible.
  3. Would be the least expensive to build in the first place.

To estimate the relative cost of power, assume this civilization has easy access to hydrogen and/or helium fusion and solar power technology. Also assume they can convert power to propulsion at approximately a 1:1 ratio. Also assume this civilization can not violate the Conservation of Mass and Energy, Thermodynamics, or other such well established scientific principles.

  • $\begingroup$ Please clarify a bit more about your setting. Do you have an ocean inside of an O'Neil cylinder? A large lake? Tidal rivers? More specifically, am I right in assuming you want to create artificial tides for creatures adapted to low and high tide conditions. $\endgroup$
    – Dragongeek
    Jul 21, 2020 at 16:12
  • $\begingroup$ Assume the builders attempted to simulate all of Earth's biomes throughout the cylinder; so, there should be all three; though, admittedly, the Ocean biomes will not be able to be supper deep; so, let's assume they could only simulate the 60m depth of continental shelf biomes and no deep seas environments exist. $\endgroup$
    – Nosajimiki
    Jul 21, 2020 at 16:24

5 Answers 5


I'm wondering how much energy it would cost to just ebb + flow the spin of the whole cylinder once per day.

Let's see:

The moon is $7.34 \times 10^{22}$ kg in mass, and perigee is 362,300 km from the center of the Earth. Deduct the radius of the Earth (6,371 km) to get the "feel" at the surface and $a = {{GM} \over {r^2}}$ the moon exerts a change of ${{6.67 \times 10^{-11} 7.34 \times 10^{22}} \over ({355,929,000})^2 } = 3.86 \times 10^{-5}$ m/s/s. Or, in context of centrifugal spin producing 9.8 m/s/s of "force", a 0.000004% change in cylinder spin.

Now, you actually get the drop for free. You can use regenerative braking, braking the rotational energy into a flywheel, that you can then re-provide during spin-up. You could design-in the friction drop to occur smoothly over the course of 12-hours, and then computer control the spin-up over the next 12-hours.

As El Duderino pointed out in the comments, having a global change in the "centrifugal" force won't produce a tide. Tides are caused by uneven dynamic forces (tidal effect is weakest directly under the moon, a spot that moves constantly as the world turns).

I wonder if this can still work, however, with the radial part of the acceleration causing clockwise and counter-clockwise "tides"; helped a little by the reduction (or increase) in the ground-pulling force.

  • $ E = mgh = \sqrt{(ma_{\perp}h)^2 + ({{1 \over 2} mv_{\parallel}^2})^2}$
  • $a_{\parallel} = \alpha r$
  • $a_{\perp} = {\omega}^2 r = {\alpha t}^2 r$

The mass of the ocean and the height don't change, so m and h can be dropped. The change in g we're looking for is $3.86 \times 10^{-5}$

  • $3.86 \times 10^{-5} = \sqrt{({\alpha t}^2 r)^2 + ({{1 \over 2} ({\alpha r t})^2)^2}}$

You can still use regenerative braking to catch most of the braking energy for recycling into spinning back up. However, it's not quite like tides work on Earth.


  • Consumes the least possible energy and resources : costs you nothing you weren't going to have to spend already to keep the cylinder spinning (although perhaps you had even lower-friction spinning; maybe costs you that).
  • Can be built in as many kinds of places as possible : the "tidal" effect is global; merely need to add water.
  • Would be the least expensive to build in the first place : only costs a few lines of software putting some timing into the re-spin schedule.
  • $\begingroup$ When you say ebb and flow you mean altering cylinder spin speed? You would brake it then spin it back up? $\endgroup$
    – Willk
    Jul 21, 2020 at 16:34
  • $\begingroup$ Unless it's frictionless, it will brake itself. Add braking, if necessary. Otherwise, just schedule how much re-spinning is done. $\endgroup$ Jul 21, 2020 at 17:04
  • $\begingroup$ There is also the issue of erosion: the constant friction of the water against one side of the O'Neill section will eventually wear it away, no matter how hard the stone, even if it is some kind of supercrystal. Structural failure thus is inevitable. But what is the intended MTTF of this structure? Thousands of years? Hundreds of thousands? $\endgroup$
    – llywrch
    Jul 21, 2020 at 17:53
  • 2
    $\begingroup$ you can slow and accellerate rotation pretty losslessly with flywheels in case natural slowdown isn't enough. $\endgroup$
    – ths
    Jul 21, 2020 at 20:00
  • $\begingroup$ @James McLellan Friction against what exactly? Isn't this cylinder floating in empty space? I highly doubt that interplanetary dust would give enough friction to have the requested cycle occur in any reasonable timeframe... $\endgroup$
    – Mr47
    Jul 22, 2020 at 8:15

You need currents for two main purpose:

  • oxygenate the water
  • recirculate the nutrients

Waterfalls and artificial rain can help you with the first one, while thermal circulation can help you with the second and also in distributing the oxygenated waters. You can use the dump heat from the activities hosted in the cylinder to create a density differential in the water bodies, which then following the gravity will induce flow in the water itself.

  • $\begingroup$ Can you explain how thermal circulation would work to this end in an O'Neill Cylinder specifically? $\endgroup$
    – Nosajimiki
    Jul 21, 2020 at 16:12
  • $\begingroup$ Imagine an olimpic swimming pool with a heater bar on the floor of the center lane. If the thermal differential is big enough, the water over this lane would flows upwards, so the rest of the lane's water will fill the empty room flooding from the bottom of the adjacent lanes. As the water in the surface gets away from the center, it cools and falls down. Now imagine the olimpic pool is hundreds of meters wide, and goes around all the station. You have convection current on the opposite axis of the water body (along it's width, not it's lenght). Sounds good enough for me... $\endgroup$ Jul 21, 2020 at 19:05
  • $\begingroup$ If there are pumps in the middle of the body of water that then circulate that water through the skin of the ship (or through some radiators) you get 2-for-1 thermal gradient and high capacity heat management for the ship. Same if you want to take the active route and refrigerate the centre of the pool in the opposite of @DanFernandez suggestion, since it’s easier to have activities that generate waste heat on the shore than under the sea. $\endgroup$
    – Joe Bloggs
    Jul 22, 2020 at 5:26
  • $\begingroup$ I was thinking more of passive internal equipment cooling, using the water body as a heat sink. Industrial equipment under the "main terraformed living deck", energy transport systems, tram like systems... yeah, I know, it's easier to just heatpipe all that to a radiation device and dump onto the void, but this way you can keep an internal heat and humidity level, and maybe even convection air flow from the cylinder axis to the rest of the station by passive equipment. A pair of big enough chunks of titanium are more durable than the best of our fan or pump systems... $\endgroup$ Jul 23, 2020 at 9:32

Inside the cylinder you are free to arrange the hydrography as you like, so a couple options are:

  1. Make the water run in a circuit and use wave pool mechanisms to keep it circulating.

Wave pools replicate the movement of the ocean one of two ways, depending on the size of the pool and the size of wave desired. In small wave pools, pressurized air is blown onto the surface of the water, or a paddle creates force in the water, creating small ripple-like waves. Other techniques utilize an "accordion mechanism" which opens and closes in order to suck water into its belly (opening) and push it out (closing) to cause waves. However, in high-volume wave pools, a large volume of water is quickly allowed into the far end of the pool, forcing the water to even out, generating a sizeable wave. In these large wave pools, the excess water is removed by being channeled through a return canal where it can be used again to generate another wave.

Make gentle waves that all flow in the direction of the circuit, and you will have an artificial current. The waves may be used for surfing too, which gives more options for leisure and fitness to the crew and passengers.

If you make the whole machinery run at really, really gentle levels you can even turn it into a lazy river, but that is not as much fun:

A lazy river is a water ride found in water parks, hotels, resorts, and recreation centers, which usually consists of a shallow (2½ ft. to 3½ ft.) pool that flows similarly to a river.[1] There is generally a slow current, usually just enough to allow guests to gently ride along lying on rafts. There may also be scenic elements added, such as small waterfalls on the edge of the river. Some connect or lead into swimming pools or wave pools, while others are self-contained courses that simply complete a circuit.

A torrent river, or wave river, is a related concept. Torrent rivers feature wave machines similar to those that are in wave pools; the waves then push riders (who are on rafts, as they are in a regular lazy river) around the river faster than they would be traveling in a regular lazy river.


Add a different kind of spin.

An object can have only one axis of rotation as I learned here:


But for ease of thinking let us think about 2 axes, realizing in our hard physics heart that there can be only one.


The cylinder spins along its long axis to simulate gravity. Yes, yes. Left image. Your left, not the images left.

Now let us also spin the cylinder along its short axis, like a propeller. Right image. This propeller spin is much slower than the long axis "gravity" spin. The effect of this gentle slow spin is to produce a vector force out towards the ends of the cylinder.

One could imagine this centrifugal force to simulate a change in altitude. Now it is "downhill" from the cylinder center to the ends. Water will tend to flow out towards the cylinder ends.

The cylinder ends admit more radiation than the central cylinder. Possibly a lot more. Water that reaches the end heats up and hot gases rise. The hot gases from the cylinders end will rise because of the same forces that pulled the water to the edges - gases are also pulled by this force, and hot, less massive gases will float atop cold. The evaporated water will make its way back to the relatively cool cylinder center and fall as rain.

  • $\begingroup$ how does this create tides? $\endgroup$
    – ths
    Jul 21, 2020 at 19:56
  • $\begingroup$ Worth noting that in the situation you describe the heating and cooling would create convective mixing with no extra spinning required. Relying on radiation seems a strange way to do it though. Place equipment high in waste heat near the shoreline and have heat transfer equipment (and presumably external radiators) in the centre. $\endgroup$
    – Joe Bloggs
    Jul 22, 2020 at 5:21

At a guess? Try placing the cylinder in orbit around a suitably sized planetary body at the appropriate distance and have the cylinder rotate so that each ends moves first closer to then father away from the larger body.

Other than that pumps and weirs and power driven hydro mechanics. If you can build an O'Neil cylinder electric power shouldn't be a problem


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .