I am procedurally generating planets for an open world space sandbox game. I am using a rough simulation of tectonic plates to create mountain ranges and other geological features.

A planet surface consists of several plates, which consists of tiles. The continental plates are randomly generated, with random initial elevation and drift velocities assigned.

Tile elevations are generated by multiplying an envelope ( e^{ -d^2 } ) with a function for different types of plate boundaries and relative drift velocities.

For the sake of realism, is it possible to have a land-ocean divergent plate boundary? I could not find anything online. There also doesn't seem to be any examples on Earth from what I could find. Is it safe to assume that divergent boundaries only occur between land-land and ocean-ocean plate boundaries?


Here is a rendering from the project. Red and green lines depict convergent boundaries. Land-land convergent boundaries produce mountains. Land-sea or sea-sea boundaries produces subduction zones. Blue lines denote a divergent boundary.

enter image description here

I'm going to change the simulation so that divergent boundaries happen only at land-land and ocean-ocean boundaries.

Sorry for my ignorance about geology. I know this is a crude way to simulate plate tectonics.

  • $\begingroup$ Wouldn't the lack of such a boundary be down to the low probability of a rift starting exactly at an ocean/continent boundary? E.g. if the East African Rift had happened just a bit further east, it would have been one. $\endgroup$
    – jamesqf
    Commented Jan 19, 2017 at 20:17
  • $\begingroup$ Nice rendering, I get the concept! Continental divergent boundaries could get a tiny oceanic plate wedged in. That would make sense for further development of your planet. Well spotted how continental plates have higher elevation than oceanic plates. Imagining that your ring is the crust of a planet, the crust would be thinner in the middle of the oceanic plates and thickening as it gets older. Mountains are also extending into the mantle, so the inner circle could almost be a mirror of the outer. Oceanic crust is much thinner than continental crust. $\endgroup$
    – user2821
    Commented Jan 20, 2017 at 7:49
  • $\begingroup$ @jamesqf Yes, the probability would be low and there are also mechanisms controlling when rifting can start. Often the rifting follows old sutures, as in the example of East African rifting. $\endgroup$
    – user2821
    Commented Jan 20, 2017 at 7:54
  • $\begingroup$ I remember some speculation that Iceland may have been a piece of continental plate that got stuck over the rift, due to the chemistry of the basalt. That may have been discounted now. $\endgroup$
    – haresfur
    Commented Jan 22, 2017 at 21:07

2 Answers 2


The oceanic plates are themselves formed from the divergent boundary, so probably not. Even if a new rifting occurred exactly at the boundary, the result would eventually be that the ocean floor surrounds the divergent boundary.

A very simplified model of an ocean would have a divergent boundary in the middle. This oceanic spreading center might have basaltic volcanic islands along it, and two mirrored plates on each sides, pushing towards neighboring continental or oceanic plates.

From wikipedia, as usual

I'm trying to think about any oddity that would fit your description, but can't really come up with a good example. Locally you might have the conditions in triple junctions eg. at Afar Triangle. Somehow, it could also describe parts of New Zealand and eastern Mediterranean, but it's very complex areas with more parameters controlling the rifting.

Earth is not controlled by any simple equation, its formed by 4.56 billion years of interaction between rocks, magma, life, atmosphere and even astronomic influence, so I'm not sure that divergent boundary never occurred between existing continental and oceanic plates, but at least it's very uncommon and couldn't last long.

A way to understand plate tectonics better, and maybe even a good inspiration for world building games, is to have a look at the (free) program Gplates that is used by students and scientists to simulate tectonic history. Don't worry about your ignorance, interest is more important than knowledge. The game might be a good inspiration for someone to learn more about something really awesome.

  • $\begingroup$ Thank you for your insightful answer! I will play around with Gplates for inspiration. $\endgroup$ Commented Jan 19, 2017 at 15:02
  • 1
    $\begingroup$ @JacquesNel I'm always very happy when animators, game developers and graphical designers make an attempt to make sense of the landscape. It's not just random shapes. $\endgroup$
    – user2821
    Commented Jan 20, 2017 at 7:38

You May think of this situation as following:

Consider an ocean-ocean divergent boundary and one of the oceanic plates (OP) is subducting below the continental plate (CP). Following shows such a setting.

<----(OP)----|(MOR)|----(OP)---->\\ <----CP----

where, OP= Oceanic Plate.

MOR= Mid Oceanic Ridge.

CP= Continental Plate

\ \ = shows subduction

--> direction of movement of plate enter image description here

Now think if the spreading ridge is coming closer to the CP. Which can be due to the difference in speed of motion of the CP. One of such scenario is also illustrated in (book William Lowrie) enter image description here

In such a scenario, though for a small duration, you can have such situation.

  • $\begingroup$ How would the spreading ridge get closer to the continental plate? Please develop your answer. $\endgroup$
    – user2821
    Commented Jan 23, 2017 at 9:23
  • $\begingroup$ The continental plate gets pushed over it as has happened to the north american plate and the east pacific (juan de fuca) plate. Once the continental plate overruns the oceanic weirds stuff happens becasue the oceanic plate is too young and hot to just sink down so you it gets layered on the bottom of the continental plate. $\endgroup$
    – John
    Commented Jan 23, 2017 at 16:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.