My world has multiple layers of floating islands, which by itself causes a lot of climatic-involved problems for me to sort out. That being said I stomped into a very simple yet important factor to take into account when designing a planet: wind currents.

Now, in a world in which the sky is full of giant floating continents and diverse pieces of land, traditional Earth-like currents can't take place (for what I know at least), since the cycle moving air from the surface to the tropopause would be interrupted by the islands.

My question is: Could wind currents behave in a way similar to that of ocean currents for each layer of islands without changing the way climate works too much?

Meaning air would move throughout the gaps between islands semi-independently from the air in the rest of the layers while still having some connection with every simple layer in orther to keep the cycle of the world intact.

  • $\begingroup$ How big are these islands? $\endgroup$
    – L.Dutch
    Jul 29, 2018 at 10:12
  • 1
    $\begingroup$ The islands in the bottom layer are of a considerable size (7 islands the each one the size of Australia aprox.) The middle layer is made up of 4 continental islands and the top layer has more disperse smaller islands (6 islands a bit smaller than the bottom layer ones) $\endgroup$ Jul 29, 2018 at 10:17
  • $\begingroup$ They're floating... are they fixed in place or do they migrate around the world on the winds? This could be a very interesting question. BTW, when you respond, if you type @username the system will automatically inform the other person of your comment. The system will auto-complete to show you which names you can use (only people who have previously commented can be notified this way). $\endgroup$
    – JBH
    Jul 29, 2018 at 22:40
  • $\begingroup$ @L.Dutch, El Nitromante answered your question but didn't know how to notify you. I'm using this as an example to show him how. $\endgroup$
    – JBH
    Jul 29, 2018 at 22:40
  • $\begingroup$ Thanks for helping me with that @JBH . Having the islands move around could make for some interesting stuff, but could also cause a series of problems that I wouldn't want, like continental crashing. My plan was to keep them orbiting the worls while maintaining partially static (they can drift a couple of meters or kilometers, depending on the island's size, but will never go x distance away from a point relative to their designated position in the map. This can, however, change due to a catastrophic event of some sort). $\endgroup$ Jul 30, 2018 at 12:34

1 Answer 1


The process of envisioning planetary weather is generally the same for all worlds. Let's begin by ignoring your continents. We'll just have a ball of (I assume) water.

  • As the sun strikes the planetary face closest to it the water and atmosphere heat up. This increases the water vapor in the atmosphere, but it also thins it out (forces the atoms and molecules to separate from each other). On the cool back-side the atoms and molecules are coming back together. So, a high pressure zone at the point closest to the sun, a low pressure zone furthest from the sun.

  • Broiling weather (aka wind and storms) forms around the sunlight terminus. This is a rotation between the heavier air near the surface and the cooler air high up, between the high pressure warm air of the sun side and the low pressure cool air of the darkside.

  • As the planet rotates, cold air from the back is brought around front and warm air from the front is taken around back. This sets up your (in the case of Earth) prevailing westerlies. Wind patterns based on latitude begin to form due to axial tilt and the fact that a small rotation with all-storm (little to no separation between the light terminator) is near the poles vs. a long rotation with little storm along the circumfrence (storms only at the light terminus and no where else, basically) at the equator. Winds due to rotation and due to the cycle between pressure zones do allow storms to be drag along their paths.

  • The cool thing about this stage of understanding basic planetary weather is that you start seeing air moving backwards against the rotation due to pressure zones colliding, storm circulation, the water vapor charge/discharge cycle, etc.

  • Finally, colder toward the poles and hotter toward the equator means your high-altitude polar cycles can form.

You didn't mention any crustial landmasses so I've assumed none. If this is the case, then you have pretty straight forward water currents that follow the rotation of the planet, and powered by the dayside-nightside hot/cold engine and latitudinal zones. AKA, they would look a lot like the wind. If you have crustial landmasses, that breaks up the possible flow of water to create currents around those masses. This creates gyres, which are important for keeping the water mixed up. This is important because...

  • Now your floating continenets and islands come into play. The air must move around them. This causes high pressure zones to form at the leading edges of the landmasses and low pressure zones to form at the trailing edges. Things start getting really complex here because that air moves at different speeds due to rotation and the amount of turbulence caused at the light termintor. At least, it's pretty complex in the real world.

  • Because in your world the air can also move beneath the land masses, which I assume are bumpy like the topside. Have you ever seen videos of air moving over an airplane wing? Right! You're going to have high pressure and low pressure zones developing on top and on bottom depending on how the landmass thickens and thins along the path of the air. Depending on how strong these zones become, they could lift the landmass higher in the air (mostly high pressure under the landmass, low pressure above) or pull the landmass lower toward the sea (mostly low below, high above). There can be raging storms below, but not above, vice-verse, both or none.

  • Then comes where your landmasses are anchored in relation to the sun. A large equatorial landmass will make a serious dent in the amount of water that evaporates during its time in sunlight. High pressure will tend to form on top, low on the bottom, meaning air's being sucked into a bit of a donut shape top-to-bottom. What heated air is taken in from the surrounding exposed water will want to drain beneath the landmass (justification for aquifers!) but will tend only to rain above if there are substantial enough mountains to block the air, creating local high/low presure zones where the air can mix and drop the water. But where it forms lakes and rivers on top, it forms drains on the bottom, like water decanting off the tip of your finger. You'd have water draining off "inverse mountain tops" into the sea to varying degrees.

  • Finally, let's bring in the "path of least resistance" rule. Air (like all gases and fluids) wants to go where it can find a state of equilibrium. This means it wants to rush from high pressure to low pressure. Think of a pressurized tank with a valved tube connecting a low pressure tank. Turn the valve and the air rushes into the low pressure tank until the pressure is equal. what this means for you is that as the air moves around two landmasses, the air between the landmasses wants to get past them and "back into the open." But, what's "the open?" The next set of landmasses create new channels for air flow. Is the first set two large north/south landmasses and the next a southern landmass and northern islands and archipeligos. Where does the air want to go? You bet, through the islands!

At this point we enter the complexity that requires computer modeling that's still regularly wrong. But, it should give you an idea of how to judge where storms and deserts will form. Think of that landless water planet and add your landmasses. Where's the air going to hit (turbulence). Where will the presure release (more turbulence)? Where are the channels? Where are the rotations? It's work, but you'll get results.

  • $\begingroup$ You are awesome @JBH $\endgroup$ Jul 30, 2018 at 15:23

You must log in to answer this question.

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