WIP- unnamed fantasy world

[ascanius]

Thanks groovey got rid of the thumbnail. The only problem with gplates is there is a learning curve and its very time consuming. What I do like however is that running 8 plate movements through time I'm starting to see the logical places where plates will form along with the places that simply don't make sense. I've had to delete a few plates already because there was no explanation on they formed. They looked good at the beginning and they made sense but once I simulated the movements of the plates I realized they're location and movement was impossible. It's a nifty program if I was confident in my python programing I would totally create an extension that simplifies things a ton.

Edit: Does anyone know if it's possible for two plates to fuse into one?

[Pixie]

Yes, plates fuse (and then re-detach sometimes) all the time. The places where they fuse are called "sutures" and normally mountain ranges are left as remnants of the period of fusing. The Urals are one example, the Carpathians another.

I am getting curious about the result of your messing about with gplates. (Haven't had the time yet to explore it myself)..

[ascanius]

Thanks Pixie, I thought it was possible but I wasn't sure. In the end I scrapped the idea of fusing two plates and left an area as it was, still not sure if it worked.

Here is the final result, hopefully. Getting the hundreds of line on gplates organized and figured out then exported to gimp took hours. The map is basic, taken from orthonographic projection so it looks skewed but on a globe it looks much better.



I don't know I'm not too convinced with how it turned out. Plate 4 was originally part of plate 2 but I couldn't think of a way that it could still be attached by that little sliver between plate 1 and 6 so I broke it up. Though now I have no idea how it is moving nor where what type of boundaries it has it is the same age more or less as plate one, plate 6 is younger. So by my thinking plate 4 should be subducted beneath plate 6 and 8. But that leaves the boundary with plate 1 and I'm at a loss.

Plate 15 and 13 were originally one plate but the continent on plate 14 broke it in two. This is one of the areas I was having trouble with. In a few spots I had a divergent boundary form on a continental plate (underwater) then later (about 100 M yrs) the continental plate would push towards the divergent boundary. As an example: Plate 13 and 15 were part of plate 14 which was moving away from the divergent boundary. Later due to movements of on the other side of plate 14 it got pushed back towards the divergent boundary. I really didn't want go back and redo half the work I already did so I decided that plate 14 broke into that of plate 13 and 15 and the heavier (older) portions of plate 14 sank beneath the lighter (younger) plate of 13 and 15 until the continent neared the boundary and broke them into plates 13 and 15. I did the same thing with plates 6 and 2.

Then there is the zone including plates 5, 16 and 17. originally the way I had plate 5 moving it sheared across plate 16 and due to it's rotation pulled away from plate 16 leaving an empty space where there was no plate what so ever, that I knew was impossible. I redid this area 3 times and finally decided to tweak the direction a little added a new plate that ends up being completely subducted by plate 14 (yeah about an hour of pointless work) then added another plate 17 to fix the problem. The main thing about these plates was I was trying to not do what I did with plates 15, 13 and 14 above. plate 5 already has very little plate wise and is mostly continent, originally the entire east coast of plate 5 was a boundary thought it would be better if I didn't have a floating continent, yet the end result is not much better.

here are two images from gplates
time 0.0


time 30 M yrs ago

[su_liam]

Hey Pixie, Thanks for your help. Sadly I've been trying to figure out gplates..... and well I've got a whole new map.

A new map is a good thing!

It's a really interesting program, don't know if you've checked it out yet but I say give it a try.

I concur.

My compliments on the tutorial, your blog helped a ton. The problem I was having with the rot files was a simple one. My text editor deemed rot extensions beneath it's lofty notice and was adamant about deciding which extension to use. In the end I used the humble notepad.

Glad it helped.

Not sure if you know but in the latest version of gplates you can now create and modify the rot file in the program without having to use a text editor, very handy. Do you know if it matters if you save a feature as unkown or fault or coastline? So far it doesn't seem to change anything.

There are a few things like poles or motion paths that have specifically implemented behaviors, but most things are merely informational for the user. That's not to say special behaviors might not be implemented in future versions...

I've had to delete a few plates already because there was no explanation on they formed. They looked good at the beginning and they made sense but once I simulated the movements of the plates I realized they're location and movement was impossible.

How so? There are quite a few seemingly impossible things that crop up, even in animations of real-world tectonic motions. Continents overrunning each other and the like. These could be artifacts of the interpolation or lack of sufficient temporal resolution at critical times when plates are sliding past each other. It could also be unremarked events where small fragments of continental crust sutured together after the close pass. Could also be continental fragments that broke off of one or the other plate in the encounter. The occasional weird, hard to explain event is part of the fun, and pretty realistic. Delete most of them, if there are a lot, but leave one or two, and try to work out an explanation. Now try to figure out how your explanation would effect the terrain.

It's a nifty program if I was confident in my python programing I would totally create an extension that simplifies things a ton.

What sort of extension are you looking at? Tectonic autogeneration of terrain has been done, but there's definitely room for improvement. Carl Davidson has a nice web app in javascript to simulate tectonic motion on a sphere, here. It takes a pretty high-end browser, because it takes advantage of WebGL for client-side processing. Not perfect, but the best of the lot. I know of a couple of similar python apps. One, by the same author as the previous app, is understandably very similar in approach, but it uses python, which you might find instructive. Another, also in python, uses a somewhat different approach, which might be similarly instructive. It also looks like, in some ways, a more interesting approach.

I can't vouch for the python apps, 'cause both of them fail on my computer for one reason or another. YMMV. Tectonics.js, clearly, isn't in python, but it works pretty well, and the javascript code with some glsl mixed in should still be pretty useful to peruse. GLSL is a lot like C(surprise!) and much of js is somewhat similar to python.

I'm not real versed in js or opengl, but I have been able to get a slightly modified version working on localhost. Fun...?

Edit: Does anyone know if it's possible for two plates to fuse into one?

As has already been mentioned, yup. They also split apart again. Not necessarily along the same lines. Laurasia and Gondwana fused to become Pangaea. Pangaea went on to fission into the current collection of continents. North America used to be part of Laurasia along with Europe, and South America used to be connected to Africa as part of Gondwana. India, Australia and Antarctica also used to be part of Gondwana, and India is now fusing with Asia, a former part of Laurasia. Infinite Diversity in Infinite Combinations.

Edit:
This thing took several hours to write overnight and when I posted it, it disappeared into expired token hell.
Hopefully this version managed to cover everything I originally wrote, but I doubt if it's half as clear or informative. My apologies if this is a confusing hash. I really need to start writing my posts offline in TextEdit...

[ascanius]

How so? There are quite a few seemingly impossible things that crop up, even in animations of real-world tectonic motions. Continents overrunning each other and the like. These could be artifacts of the interpolation or lack of sufficient temporal resolution at critical times when plates are sliding past each other. It could also be unremarked events where small fragments of continental crust sutured together after the close pass. Could also be continental fragments that broke off of one or the other plate in the encounter. The occasional weird, hard to explain event is part of the fun, and pretty realistic. Delete most of them, if there are a lot, but leave one or two, and try to work out an explanation. Now try to figure out how your explanation would effect the terrain.

Maybe not impossible but more problematic. In one case I had a plate cut across the divergent boundary of another. This posed a problem because I needed that divergent boundary to explain the plate motion. In the end I deleted the plate and added two smaller ones in it's stead. One of which disappears.

What sort of extension are you looking at? Tectonic autogeneration of terrain has been done, but there's definitely room for improvement. Carl Davidson has a nice web app in javascript to simulate tectonic motion on a sphere, here. It takes a pretty high-end browser, because it takes advantage of WebGL for client-side processing. Not perfect, but the best of the lot. I know of a couple of similar python apps. One, by the same author as the previous app, is understandably very similar in approach, but it uses python, which you might find instructive. Another, also in python, uses a somewhat different approach, which might be similarly instructive. It also looks like, in some ways, a more interesting approach.

Thanks I wasn't aware of these. I checked them out and gonna have to play around with a few and see what I can come up with.

As has already been mentioned, yup. They also split apart again. Not necessarily along the same lines. Laurasia and Gondwana fused to become Pangaea. Pangaea went on to fission into the current collection of continents. North America used to be part of Laurasia along with Europe, and South America used to be connected to Africa as part of Gondwana. India, Australia and Antarctica also used to be part of Gondwana, and India is now fusing with Asia, a former part of Laurasia. Infinite Diversity in Infinite Combinations.

I added two suture zones one I very large and long the other smaller. Hope I didn't over do it.

Edit:
This thing took several hours to write overnight and when I posted it, it disappeared into expired token hell.
Hopefully this version managed to cover everything I originally wrote, but I doubt if it's half as clear or informative. My apologies if this is a confusing hash. I really need to start writing my posts offline in TextEdit...

LOL. Happens to me all the time, I've gotten very good at copying the message before I press submit reply.....Like it just happened now.


Now here is my latest tectonic map with the continents. I haven't yet placed most island chains and I need to tweak the coastlines a bit and reduce the number of islands in some areas. Overall I am pleased there are a few things I might change but I'm going to try to keep this without doing any major changes unless needed.

Blue= subduction, side with the circles is the one being subducted, I still get this confused.
Red= divergent
Greed= transform
Purple= convergent
Purple shadow= orogeny



I tried to give a general idea of the plate rotation with the curved arrows. What do you guys think?

[su_liam]

I've been pretty good about copying posts before I send them out. A habit I picked up from making monster posts on word press . Unfortunately, I've been forgetful, lately. I had to redo the last third of my last blog post, too . Early-onset dementia, I guess...

One thing that didn't survive my lost post was an idea for a python extension to gplates that would be very welcome. If you could feed in an existing Euler Rotation along with a bearing, distance and rotation on the surface as a delta. it could output a new Euler Rotation. This could be implemented either as an extension to gplates or as an offline app. Bonus points if it can read in a file with a series of rows with times, step distances, bearings and rotations along with plate ids and output a rot-file. More bonus points if it could be used to extend an existing rot-file!

I have a few observations on your map, if you're interested.

Plate 4 appears to be a dying remnant plate. RIP. Should probably be surrounded by island arcs.

The boundary between 1 and 3. Mid-ocean ridge converts into a subduction trench? That seems a little dubious. The boundary between 5 and 3 is similar.

26. I like this little plate, but should its southeast boundary be convergent? I'm just not sure.

It seems like 13, 15 and 16 were formerly a single large plate that is being devoured by plate 14 with some help from 18. I like this, it gives a feeling of history...

The ocean boundary between 2 and 6 seems like it should be transform.

Generally, I think you need more island(or seamount) production on the overthrusting slab of subduction boundaries. Like the ones surrounding plate 4...

[ascanius]

I've been pretty good about copying posts before I send them out. A habit I picked up from making monster posts on word press . Unfortunately, I've been forgetful, lately. I had to redo the last third of my last blog post, too . Early-onset dementia, I guess...

Lol, do sudoku.

One thing that didn't survive my lost post was an idea for a python extension to gplates that would be very welcome. If you could feed in an existing Euler Rotation along with a bearing, distance and rotation on the surface as a delta. it could output a new Euler Rotation. This could be implemented either as an extension to gplates or as an offline app. Bonus points if it can read in a file with a series of rows with times, step distances, bearings and rotations along with plate ids and output a rot-file. More bonus points if it could be used to extend an existing rot-file!

I don't understand what you mean. Do you mean taking the change in plate vector at time initial and time x to get the vector of time y? Or do you mean to use the change in rate to extrapolate plate vectors? Or I think this is what you mean, take the change of vectors over time to calculate the euler pole that would be used, instead of having to set it yourself. I do agree that would be continent. I would need to really brush up on my trig to do the last one.

I have a few observations on your map, if you're interested.

Plate 4 appears to be a dying remnant plate. RIP. Should probably be surrounded by island arcs.

Yeah, I've added island arcs.

The boundary between 1 and 3. Mid-ocean ridge converts into a subduction trench? That seems a little dubious. The boundary between 5 and 3 is similar.

Time for interesting explanations. I did the plates in gplate and what happened is plate 1 and 3 split apart and that divergent boundary ran down to plate 11. Do to the plate rotation, mostly the rate at which new crust is being along the eastern divergent boundary between 1 and 3, the two plates are pushed back together at that subduction zone. That divergent boundary is slowly being closed. With 5 and 3 you get the same thing along the northern portion the plates are pushing together in such a way to close the divergent boundary. Whether or not it works I don't know, I'll have to take a look at it and see if I can find a better solution.

26. I like this little plate, but should its southeast boundary be convergent? I'm just not sure.

With plate 26, because it doesn't have a divergent boundary anywhere I have do idea which direction it would be going. I sorta just drew an arrow and forgot to go back and see how it works. I was thinking of averaging the direction of plates seven and three for the direction of 26. Or maybe have a more westward direction? I mean there is a lot of land mass pushing it more westerly and much less pushing it to the north so maybe matching direction with 3 would make the most sense.

It seems like 13, 15 and 16 were formerly a single large plate that is being devoured by plate 14 with some help from 18. I like this, it gives a feeling of history...

Thanks, happened entirely by accident. However if you look closely plate 11 is doing the same with 8 and 9, while plate 3 is doing the same with plates 7 and 9. I'll be honest and say that these were by accident. In gplates I noticed that I had two voids where plates 8,12,7,9 were and realized I needed a divergent boundary somewhere along there.

The ocean boundary between 2 and 6 seems like it should be transform.

For this I wasn't sure what to put. I think I'm going to go back and create a mix. Plate 6 is moving at a faster rate then plate 2 so I think it would offset the transform boundary. hmmmm. That or changed the general direction of plate 6.

Thanks for the help, gonna go back and see what I can do to fix these problem areas.

I'm having trouble getting the map to upload I'll try again later.

[Pixie]

Hey, great improvements - looks like g.plates provided you with a really successful crash course on tectonics.

Apart from the occasional divergent-oceanic-boundary-turns-into-subduction-area which is impossible to explain, everything looks fine and plausible and you have a wealth of unique geographical places, like the northern side of plate 14 (it seems like a boundary where trench rollback has been "rolling back" for millions of years, or like the micro plates 21 to 24, where it really feels like there used to be a plate there but it's just cracking up into pieces, or like the probably very important for trade strait between 10 and 3.

Now, as for things I would do differently (but, please, consider all this with a pinch of salt, I'm no guru):
- reconfigure the microplate north of plate 5, where it meets plate 3 - it's ocean crust vs. ocean crust, so you can't have the plate on top shutting down the magma outflow; maybe one extra microplate, rotating, could fix it.
- review curvature of boundary between plates 2 and 6, make some changes on the type of boundary; it looks like it could be transform most of the lenghth.
- enlarge plate 7 a little bit to the north, and make those subduction boundaries more curved; or alternatively, break that southern tip of plate 1 away from the main body, it could be another "nearly dead" plate like 4 or it could be fueled by a northern divergent boundary (but almost magma depleted)

On a different subject, and I know this map of yours is still in draft form when it comes to looks. It took me a while to understand the colors at the coast (if I did) - my understanding is that you have two levels of underwater depth shown. Is this right? That gray isn't helping, as it is darker than the colors on either side.

[groovey]

The coastlines and the shelves look yummy (doesn't the land mass under the nº3 remind anyone of a fish? Lovely).

The only thing I could make a minor issue of, is that the subduction circles are very small, even when you zoom in the image. As they are they get the job done to indicate the side of subduction, but only when zooming in.

I'm sorry to hear about your wrist, hope you get well really soon and can continue working. I hope it's nothing too serious.

[Pixie]

The only thing I could make a minor issue of, is that the subduction circles are very small, even when you zoom in the image. As they are they get the job done to indicate the side of subduction, but only when zooming in.

I second this, larger circles (or triangles) would be better.