Using a Fractal Generator, I generated the attached map. I really like the shapes of the Continents & Islands, and want to preserve as much as possible. But, I also want realism, and realize I need to overhaul their interiors (to add mountains, & accurately model climates).

What I'm asking is for is a critique of my sketch of the Plates. I want to know if the design could realistically produce the landforms of the original Fractal Map, including the Archipelagos. I'm having trouble understanding the Geology, specifically which islands would need to be Volcanic, and if I need to add more Plates.

Any additional suggestion or critique would be greatly appreciated. Since this map is a separating Supercontinent, knowing where old mountain ranges need to be located (and how high they can be) would be nice.

{Edit: I was dreading this... I have crudely attempted to draw which directions I think the plates of concern should be heading in, and numbered them (Most concerned about the numbered plates, particularly 1 & 2, as well as 3). Above all, its that archipelago, off the East Coast of the Central Continent, that I'm concerned about. I'll try to figure out the rest myself, but I want those Eastern Island Chains to exist, as they're Jungle Islands (For Possible Pirate Adventures)}

enter image description here enter image description here

  • $\begingroup$ you need to differentiate between convergent and divergent margins. $\endgroup$
    – L.Dutch
    Commented Feb 24 at 5:25
  • $\begingroup$ there is some problems, plane boundaries mostly fall on the land portion of a coastline or in the middle of the ocean, the boundary wrapping around the outside of your complex of continents (W, S, and NE) should not be there. the map as a whole is not too bad. Your continental outlines should also fit together for most of your intercontinetal margins. $\endgroup$
    – John
    Commented Feb 24 at 14:54
  • $\begingroup$ this would be a lot easier if you numbered the boundaries so we can easily tell you which ones are unrealistic. instead of "the eastern most continental margin" $\endgroup$
    – John
    Commented Feb 24 at 14:58
  • $\begingroup$ I added some margins, as it seems that was the actual thing I should have been asking. The interactions between plates 1, 2 & 3 are what I'm most concerned about. $\endgroup$ Commented Feb 24 at 16:56
  • $\begingroup$ The next thing to think about is the rate of all those margins. For example, 1 has a right push on one side as well as a left push on the other side. The rate of those two will determine a lot. Plate 3 seems to be solid and moving but the rates on the edges will make a difference. There should be something happening on the boundary between 2 & 3. $\endgroup$
    – David R
    Commented Feb 25 at 15:01

1 Answer 1



If we look at real-world plate tectonics, convergent plates such at the main northern plate and the plate north of plates 1 and 2 that is moving northward should produce a mountain range around the plate boundary as the lighter continental rock piles up as the plates collide. Even if the northernmost plate is subducting beneath the more southerly plate, the mountain forms don't match. In such a case, we would expect a long, narrow range of volcanic mountains such as we see in Chile, South America, with the Nazca plate subducting beneath the South American plate.

The whole problem is that this map was fractally generated. Fractal generators don't usually take plate tectonics into consideration, so there are no mountain ranges that could have been caused by tectonic processes, only arbitrarily-placed land masses and high points.

If we look at real-world geographic maps, we don't generally see land masses full of mountains, the mountains tend to occur on plates above the places where an oceanic plate subducts beneath a land plate, such as the Pacific Plate subducting beneath the North American Plate and causing the west-coast mountains, or the aforementioned Nazca/South American subduction zone.

Alternatively, a collision zone such as the Indian Plate and the Eurasian Plate may cause a high mountain range as the lighter continental rocks pile up as the heavier plate beneath subducts beneath the other continental plate.

If you want a world with plate tectonics, it's best to do so yourself, from first principles, by hand, since I've never found a decent global plate tectonics simulation that will work with a non-earth world.

You need to start with plume theory, where plumes of hot rocks rise from the planet's core and as they spread, pull along the thin plates above them. The plates spread out from several locations, and are pushed along to collide with other plates between those plumes, where colder (but still molten) rock descends again.

Once you know where your plumes are (typically 4 or 6 plumes) and where your plates are, you can chuck some light rocks on top and see where they go, remembering that while plates subduct, the shield rocks on top don't tend to vanish. You get volcanism where a plate descends below another, and the friction melts the rock and causes volcanoes above the descending plate... which may form islands or mountain ranges, depending upon their age.

Add transform faults that split land masses apart but don't cause mountains, such as Baja California, and you have an idea of how realistic mountains are created. Add erosion to reduce the size of mountains, create valleys and river deltas, and that's most of the things you need to know... aside from meteoric impact craters.

Leave the fractal generators alone, they don't do you any favours.


You must log in to answer this question.

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