Pangarap: Building a world from the ground up

[Tiluchi]

Zoinks Charerg! This is fantastic, and thanks for putting all this work into my project- I hope it comes in handy with your new projects. I've been doing some of my own fiddling around with geography and plates, and I'll post some of that soon, but suffice it to say that I've found decent ways of altering the geography to fit most of your suggestions. Anyway, I just wanted to say thanks again! I'll be incorporating these into my map and plate setup as much as I can.

[Charerg]

Zoinks Charerg! This is fantastic, and thanks for putting all this work into my project- I hope it comes in handy with your new projects. I've been doing some of my own fiddling around with geography and plates, and I'll post some of that soon, but suffice it to say that I've found decent ways of altering the geography to fit most of your suggestions. Anyway, I just wanted to say thanks again! I'll be incorporating these into my map and plate setup as much as I can.

You know Tiluchi, with those words you've just volunteered your world for some hardcore tectonic modelling . Just kidding, you can choose to incorporate as much as you want. Anyway, I think this will be useful both for myself in testing out a few of my ideas with GPlates, and also for any tectonics junkies out there who might find this stuff educative.

Right, so I actually developed the model a bit further. The positions and rotations of the continents remain the same as in the last version. However, I added Ilang to the puzzle, and made a pretty ambitious attempt to model the subduction of a mid-oceanic ridge with flowlines and a "ghost plate" representing the subducted crust (it worked out pretty well, too).

So, here's the GIF of this version (I guess this could be called more-or-less final):
Pangarap_IV_Breakup.gif

Since I ended up adding quite a few things I want to comment on, I'll probably have to split this into two posts (because the whole explanation would be too much of a wall-of-text even for me).

Okay, first off, the arrangement of the SSc 200 mya (Ilang was a fairly natural fit, as it happened):
200 Mya.png

So, due to Ilang's position the Abaca-Ipil block can't detach without Ilang detaching as well, so that is what I envision happened. Here's the situation 170 Mya:
170 Mya.png

I also added some volcanic action into the mix. The black areas represent a LIP, while the red dot is an associated hot spot (which will be tracked from 180-80 Mya in this model). Btw, it's very easy to track hotspot trails with motion paths in GPlates.

Next up, 150 Mya:
150 Mya.png

Here, the main rift jumps to the north of Ilang (think Madagascar detaching from India), breaking off a chunk from the northernmost portion. If you're still thinking about that India-style collision, there could easily be a larger chunk here that breaks off and eventually collides with Molave (though that would be a past collision during the present-day map).

And 125 Mya:
125 Mya.png

This should be fairly self-explanatory. In this model, the Abaca-Ipil Block and the plate north of Ilang move at slightly different speeds, so there's a transform fault in between.

100 Mya:
100 Mya.png

Around this date the Abaca-Ipil block collides with Malobago, and whatever pieces broke off Ilang probably collide with Molave. The subduction zone reverts, and the trail of oceanic crust (forming what I've dubbed the "Titanic Plate") starts to be subducted.

I think all this should be fairly clear so far, so I'll leave it at that for this post. To be continued (probably tomorrow)...

EDIT:
Oh, I should mention that present-day Ilang changed location in this version. Otherwise the continent arrangement is the same as in my previous "tectonics sketch".

[Tiluchi]

Again Charerg, all I can say is wow! Fantastic work, and thanks so much for going through all of this. It also looks as if you're using a newer version of GPlates than I have, since I can't find some of the tools you seem to be using here, such as the "Motion paths" feature, or the aging of the oceanic crust.

Anyway, I figured I'd jump in here with my latest updates before you have me inventing a new continent or something Overall, I've followed most of your suggestions, with a couple of exceptions. First, I liked your idea of having Narra split off from from the southern supercontinent (I'd had that idea earlier, but couldn't make it work with the previous direction of Malobago). However, I decided to go a step further and have already collided with Malobago, so I could have an excuse for some nice high Himalaya-style mountains, which Pangarap previously lacked.

Second, rather than having Ilang split off from Talisay, I've moved it west and had it split off from the east coast of Dao, somewhat after the Ipil-Abaca complex and Narra did their own rifting. That allowed me to have it creeping in southeast of Narra, and giving me the nice complex volcanic island chain structures I wanted, as well as making the continent a little more accessible for worldbuilding purposes.

I also made some other cosmetic changes, like adding a large failed rift to northwestern Malobago to give that some more interesting geography, and shifting the island chain off of eastern Molave to fit better with the subduction zone.

Here's the geography and tectonic boundaries I have so far. I intentionally left the boundary between Bakawan and Talisay rather vague, since it seems you have some more ideas on that, and I'm not particularly attached to any tectonics there yet (as long as those island chains stay there).

PangarapPlates2.jpg

A few questions I still have regarding the tectonics:

• Is the back-arc basin between Narra, Malobago and Abaca still valid even with Narra already colliding with the main continent? I'm making the assumption here that Narra is still advancing northward, so the collision hasn't quite finished yet.

• Should there be a subduction zone to the south of Ipil, or should the subduction just end where Narra is rifting away from the Malobago complex? I'm assuming the latter, since it doesn't make sense for it to be subducting on opposite ends.

• On a related note, should I just have Ipil and Molave have a subducting border with each other? At the moment there's a sort of disappearing Juan de Fuca-style plate, but I don't know if that's necessary (or a big deal).

• Does the oceanic plate between Talisay, Ipil, and Molave make sense, or should the rift just be between the Talisay and Ipil plates? Either solution looks rather awkward in my view, unless you have an alternate option (now that Ilang is on the other side of that central ocean).

Not to mention I'm sure there are other mistakes with the map. Looking forward to thoughts on the current state of the tectonics!

[Charerg]

Right, I'll comment on your post later Tiluchi, since I kind of need to write that "follow up" to my last post before I forget about what I intended to say .

So, continuing where we left off, 60 Mya:
60 Mya.png

Right, I guess this is where things start to get complicated. So, we can see the "Titanic Plate" has almost sunk into the depths and a new rift forms, breaking Narra and Ilang away. The old hot spot that we tracked from 180-80 (nicknamed "Daogal" after the continent close by) has become inactive. Instead, I've envisioned a few new "hots" appearing: one south of Narra (nicknamed "Narragal") and one close to Ilang ("Ilangal"). I should mention the Ilangal hot should actually be at the northern end of it's trail (Ilang plate has an absolute southward motion from 80-50 Mya, although it has a northwards relative motion compared to Talisay).

There's been a new LIP formed south of Ilang. At this point I might comment a bit about the rifting there. So, in my prior post the crust between Ilang and Talisay was depicted as "oceanic crust". However, rifting is often more complicated, and we might envision that perhaps it's an area of extended ("stretched") continental crust instead, with maybe a sliver of actual oceanic crust (similar to the Rockall Basin in the Atlantic). If the crust there is already weakened by the previous rifting episode, it would be likely for the "next rift" to form there as well.

Next up, 40 Mya:
40 Mya.png

Around 50 Mya, a major re-arrangement of plate boundaries occurs because the Titanic ridge is subducted. I envision the Narra-Ilang ridge falling inactive, with the Narra crust becoming a part of the Ilang plate (represented by the grey mass). Also, a new subduction zone develops east of Narra (think the subduction zones east of Australia) as part of the process, starting to subduct the ancient, loaded oceanic crust there. Btw, it's notable that the Ilang plate actually has an eastwards absolute motion in this model form 50-0 Mya. More about that later.

And 20 Mya:
20 Mya.png

So, we can see the previous processes continuing here. One thing to note is that the "Daogal Trail" is actually subducted by those "Kermadec style" subduction boundaries to the east of Narra (so tracking the whole thing was for naught! )

And finally, present day:
Pangarap IV.png

Btw, I made the oceanic crust transparent this time around, so the flowlines are more easy to see. Btw, that crust isn't automatically generated, it's just slabs of crust drawn normally (just like continents) at specific intervals, then given the right Plate ID so they follow the correct plate. This is fairly easy to do with flowlines, since they tell you where the ridge is located at any given point of time.

For changing the colours of the plates (including age-based colouring), check the GPlate tutorial on the subject. About flowlines (and motion paths work the same), I made a post a while back (and there's also a GPlates tutorial about them). I should amend my earlier post that when generating a flowline choose the spreading centres option (otherwise GPlates will crash 90% of the time). Also, save your project before attempting to create a flowline, just in case it crashes anyway.

I might try to put together a tutorial of some sort about this (when I have the time).


Okay, so about the movements of the Ilang Plate:

So, looking at that hotspot trail I though that the movements of this plate didn't look quite right. And sure enough, when checking the speeds (GPLates tut) it turns out it crawls at a measly 1.8 cm/year, which doesn't seem quite right for a plate made of only oceanic crust (and being subducted). Meanwhile Talisay is cruising around at 3.5 cm/year . Erm, even the best models have minor flaws....

I also checked the present day velocity vectors for the plate:
Ilang Plate Velocity.png

As can be seen, the movements are a bit suspect to say the least for this plate. So as an update it might be an idea to change the rotation so it moves in a more nortwest direction. This would require updating the model (might do that in a future post if this is still relevant to your updated tectonics).

Anyway, I've incorporated the "Ilang northwest drift" concept into my tectonics sketch. In this case, some of Molave-Ilang boundary would be a transform boundary instead of subduction (think North American western coast, San Andreas fault and so forth). The situation is also a bit similar with NAmerica: Molave has subducted the "Titanic ridge" in the relatively recent past, just like NAmerica subducted the ridge of the Farallon Plate (of which, Juan de Fuca, Cocos and Nazca are remnants).

The sketch:
Pangarap Plate Sketch II.png

Note that those volcanic arcs east of Narra don't necessarily have to be in this location (or have this shape). You could also change their shape, place them much closer to Narra etc. This is just a quick sketch about the general idea. Similarly I have an idea about a Caribbean-style region in the Talisay-Bakawan boundary, since the situation there is reminiscent of the boundary between the American plates. Again, this is just an initial sketch of how this area could turn out, nothing too authoritative.

So, that turned out to be a big wall of text (and pictures), but I think I managed to cover most of what I intended. I'll try to comment on your latest post somewhat later, Tiluchi.

EDIT:
I should amend my model a bit in regards to that "Kerdamadec-style" ridge. Unlike mid-oceanic ridges (which create crust at equal speeds on both sides of the plate), back-arc basins actually have asymmetric seafloor spreading, so you can't really use flowlines to depict them (more on the subject in wikipedia).

[Tiluchi]

Oh dear, I suppose I went and confused things putting Ilang on the opposite side of the ocean, didn't I? I suppose the tectonics are a bit "easier" (relative term!) with it on the eastern side of the inner ocean, but world-building wise it makes more sense to me to have it colliding with Narra. Unless I decide to go a little crazy and add another Australia-style continent whose movement mimics that of Ilang in your reconstruction. Hmmm... an idea that does have it's merits. I'll try sketching it in and see if it could work. Anyway, no worries if you don't want to redo everything just because I switched the position of one continent! I already did a quick reconstruction in GPlates and it seems to work alright.

I do rather like your idea for the Talisay-Bakawan area... I had something similar in my previous version but didn't copy it onto this one. Think I'll go ahead and rearrange some geography to make that work!

[Charerg]

Oh dear, I suppose I went and confused things putting Ilang on the opposite side of the ocean, didn't I? I suppose the tectonics are a bit "easier" (relative term!) with it on the eastern side of the inner ocean, but world-building wise it makes more sense to me to have it colliding with Narra. Unless I decide to go a little crazy and add another Australia-style continent whose movement mimics that of Ilang in your reconstruction. Hmmm... an idea that does have it's merits. I'll try sketching it in and see if it could work. Anyway, no worries if you don't want to redo everything just because I switched the position of one continent! I already did a quick reconstruction in GPlates and it seems to work alright.

I do rather like your idea for the Talisay-Bakawan area... I had something similar in my previous version but didn't copy it onto this one. Think I'll go ahead and rearrange some geography to make that work!

No worries, I don't think there is necessarily anything wrong with your idea about Ilang's position and movement. It would just require updating the model, so the SSc has a different arrangement and so forth. Like you said, I might leave you to do the updating though .


EDIT:
If you want to mimic my model considering the "Ilang Plate", here's the quick and ugly "fixed version" btw:
Pangarap IV Alt.png

The grey piece represents the approximate boundaries of the Ilang Plate (which need to be defined for velocity fields to work), and I've relocated the associated oceanic ridges using flowlines. Also, I moved the "Kermadec Zone" east of Narra much closer to it (which makes more sense). As usual, the Talisay flowlines are messed up (due to having that present-day rotation in this).

EDIT2:

And the "Ilangal Hotspot Trail" is off in the above pic. Here's what it actually is:

Ilangal Trail Zoomed.png

The trail goes initially southwards since the Ilangal Plate moves southwards with the "cruising Talisay" during initial rifting (well, according to wikipedia's map, SAmerica has about 3,2 cm/year relative to Africa, so maybe 3,5 cm/year for Talisay isn't that crazy ). After that, the Ilangal plate started moving northwest.

And that is probably sufficient development for this particular model. I'll try to post something relevant to your updated tectonics map next .

[Charerg]

Here's the geography and tectonic boundaries I have so far. I intentionally left the boundary between Bakawan and Talisay rather vague, since it seems you have some more ideas on that, and I'm not particularly attached to any tectonics there yet (as long as those island chains stay there).

PangarapPlates2.jpg

A few questions I still have regarding the tectonics:

• Is the back-arc basin between Narra, Malobago and Abaca still valid even with Narra already colliding with the main continent? I'm making the assumption here that Narra is still advancing northward, so the collision hasn't quite finished yet.

• Should there be a subduction zone to the south of Ipil, or should the subduction just end where Narra is rifting away from the Malobago complex? I'm assuming the latter, since it doesn't make sense for it to be subducting on opposite ends.

• On a related note, should I just have Ipil and Molave have a subducting border with each other? At the moment there's a sort of disappearing Juan de Fuca-style plate, but I don't know if that's necessary (or a big deal).

• Does the oceanic plate between Talisay, Ipil, and Molave make sense, or should the rift just be between the Talisay and Ipil plates? Either solution looks rather awkward in my view, unless you have an alternate option (now that Ilang is on the other side of that central ocean).

Not to mention I'm sure there are other mistakes with the map. Looking forward to thoughts on the current state of the tectonics!

Okay, so lets start with the questions.

1. I don't see any reason why not. Well, it would make more sense as an old back-arc basin, sort of like the Black Sea. It wouldn't probably be an actively expanding back-arc. But more about that collision later.

2. Yes, there should be a subduction zone south of Ipil (because there was one previously, and also the oceanic plate converges towards it). However, since Ipil is sort of diverging away from the ocean it's subducting, this is a "divergent subductive margin", and would likely develop a back-arc basin. So there should be something like the Aleutian Arc there.

From wikipedia:

Back-arc basins are hypothesized to form as a result of trench rollback (also, hinge rollback). This is the backward motion of the subduction zone relative to the motion of the plate which is being subducted. As the subduction zone and its associated trench pull backward, the overriding plate is stretched, thinning the crust which is manifest in the back-arc basin

3. Yes, the crust north of Ipil should be subducting under Molave, this is what causes Ipil to be pulled towards Molave and creates the Abaca-Ipil rift in the first place.

4. I can say that those random transform faults in seemingly arbitrary directions don't make a lot of sense. However, since this is fairly different from what I modelled, I can only guess at the actual shape and borders of the oceanic plate there. "Your guess is as good as mine" situation, I suppose. That said, making the motions/borders of the oceanic plate something like the Ilang plate in my model seems a better solution to me.


Also, about that Narra-Malobago collision, I made an educative picture. I've just drawn the tectonic borders from memory here, so there might be some minor inaccuracies.

Convergent boundaries:
Convergent Boundaries.png

As you can see, the concave shape of the Narra-Malobago convergent boundary only makes sense if it's an island arc-continent collision. If this is a continent-continent collision, it should have a convex shape on the upthrown (Malobago) side.

That's it for now. I do have some more criticism that I'll post tomorrow. Hope you find this useful !

[Charerg]

So, to continue at my comments/criticisms:

First off, it's worth thinking a bit about aulacogens and where they've formed on Earth. As you probably know, an aulacogen is a "failed arm" of a triple junction. The "arms" of a triple junction tend to have about 120 degree angle to each other (roughly). Here's a few examples:

Anza Trough (Gondwana splits into west and east Gondwana):
Anza Trough Opening.png

Benue Trough (SAmerica splits from Africa):
Benue Trough Opening.png

With those examples in mind, to me the most likely location for an aulacogen in the Malobago-Molave split is here (even more so since it's a boundary between two cratons):
180 Mya Extended Crust.png

Indeed, you have that region as an active rift in your present-day tectonics. Personally I'd say it would work better as an aulacogen rather than an active present-day rift.

Right, that's it for the tectonics-related commentary. Next I have some criticisms about the new continent shapes. Bear in mind that these are just my opinions, and everyone might not agree with them. Also, I might be somewhat attached to the old continent shapes because I'm kind of used to looking at them after doing that GPlates model.

Let's start with the eastern portions of the northern continents:
Northern Continents - East Portions.PNG

You can see directly at Map to Globe that the mid-oceanic ridge is well off the mark. Also, Molave ends rather "abruptly". Clearly the subdction zone should continue further eastwards. Similarly, those islands north of Molave look weird unless there's some explanation for them that I'm missing. They would work much better northeast of Molave, since there should be a big chunk of continental crust there. Also, I think those "3 inlets in a row" at Malobago look too artificial. Maybe fill the middle inlet or something (again, this is just my opinion).

Then, the western portions:
Northern Continents - West Portions.PNG

I feel that the "hammer shape" of western Molave looks too symmetrical (and therefore artificial). I think you should adjust it to look more irregular (and therefore natural). Also, the Abaca peninsula fits neatly in a box, and I think the shape is too regular here to look truly natural. Again, just my opinions, hopefully someone else will comment so we can get a "3rd perspective" about the continent shapes.

That's it for my criticisms. Keep up the good work and looking forward to see where this progresses .

[Tiluchi]

Thanks as always Charerg for explaining things clearly, and for the thoughts and criticisms. I agree with... most of them. I'll see if I can respond to them individually.

First of all, I liked your reconstruction of Ilang on the eastern side of the inner ocean, so even though I've moved Ilang to the western part, I've decided to add another small continent (let's call it Ilang2) in the eastern half of the ocean. This actually works rather well, as I'd like to have an analogue of both Madagascar and Australia, fauna-wise. I'm still not 100% sure about how I want the shape of the continent or its exact location to look, but I do like the addition... for now.

1. I don't see any reason why not. Well, it would make more sense as an old back-arc basin, sort of like the Black Sea. It wouldn't probably be an actively expanding back-arc. But more about that collision later.

Ok, that makes sense. Based on your thoughts I've made it a back-arc basin on the verge of going inactive, as Narra finishes its collision with Malobago proper. Of course, I'm not exactly sure what the tectonics would look like in that case since (so far as I know) there's no parallel to it on Earth. I picture the large island as something of an analogue to Taiwan, which from my understanding is basically a collection of accreted island arcs. In addition I've changed the boundary once it reaches the back-arc basin to concave rather than convex, as it seems that's more of an island-arc collision area. Let me know if that's not correct though.

2. Yes, there should be a subduction zone south of Ipil (because there was one previously, and also the oceanic plate converges towards it). However, since Ipil is sort of diverging away from the ocean it's subducting, this is a "divergent subductive margin", and would likely develop a back-arc basin. So there should be something like the Aleutian Arc there.

Ok, that makes sense. I'll admit I'm still not entirely sure where the subduction zones should lie since the whole "diversion-subduction zone" is rather... odd... as a whole.

You can see directly at Map to Globe that the mid-oceanic ridge is well off the mark.

I blame the difficulty of drawing a polar plate margin in equirectangular format! If only Photoshop had an extension that allowed you to write on a globe... I've sort of redone it, but if it's still not entirely correct it's because drawing in the poles in equirectangular makes me dizzy :p

Similarly, those islands north of Molave look weird unless there's some explanation for them that I'm missing. They would work much better northeast of Molave, since there should be a big chunk of continental crust there.

Hmm, I suppose I hadn't thought too much about the locations of continental crust, especially since the plate directions changed. I've redone the northeastern bit of Molave, moving the islands there and adding some more land so that I can extend the subduction zone. Have I mentioned that drawing geography at the poles in Equirectangular gives me a headache?

Also, I think those "3 inlets in a row" at Malobago look too artificial. Maybe fill the middle inlet or something (again, this is just my opinion).

I suppose that's fair, though I guess I should point out that in this moment the coastlines are still more or less rough estimates, and I expect them to change at least a little bit, sometimes drastically, as I figure out more of the geology and draw the coastlines on a proper scale. I've changed the geography of northwest Malobago somewhat, however, to make it a little more interesting. I suspect that will change again somewhat once I go through and redo the coastlines at full size, thinking about what would look good on a regional map.

I feel that the "hammer shape" of western Molave looks too symmetrical (and therefore artificial). I think you should adjust it to look more irregular (and therefore natural).

Oy vey, northern Malobago has been a huge source of frustration for me the past couple of days, as I try to come up with a shape that's more or less natural, but also looks decent both in map format and on a globe. I think what I have now is more or less finalized... though of course it's going to get much more fjord-y!

Also, the Abaca peninsula fits neatly in a box, and I think the shape is too regular here to look truly natural.

I see your point, but I'm also trying to train myself not to be afraid of "boring" land shapes, since lots of the landforms on Earth look rather uninteresting from a cartographer's point of view- say, Africa, most of South America, Australia, or the Arabian Peninsula (which is more or less the same size and shape as Abaca). Although again, the coastlines aren't final yet- that's a future step. That said, I did change around the shape of the peninsula a little bit to fit with the whole back-arc basin thing, so it's a little more interesting-looking now

Indeed, you have that region as an active rift in your present-day tectonics. Personally I'd say it would work better as an aulacogen rather than an active present-day rift.

Hmm, that's fair, but I also happen to really like the idea of an active rift valley! Perhaps it could be an aulacogen that reactivated due to the stress of the multiple active margins along Molave...

At any rate, here's what I have at the moment. Still have to do some more thoughts about the other geological features, and refine the coastlines and everything before I begin with the climates. But first I think I need to take a break from the tablet for a couple of days, before I start dreaming of equirectangular projections and back-arc basins!

PangarapPlates.jpg

[Charerg]

Ok, that makes sense. Based on your thoughts I've made it a back-arc basin on the verge of going inactive, as Narra finishes its collision with Malobago proper. Of course, I'm not exactly sure what the tectonics would look like in that case since (so far as I know) there's no parallel to it on Earth. I picture the large island as something of an analogue to Taiwan, which from my understanding is basically a collection of accreted island arcs. In addition I've changed the boundary once it reaches the back-arc basin to concave rather than convex, as it seems that's more of an island-arc collision area. Let me know if that's not correct though.

Sounds plausible enough to me. Tbh, I'm not 100% sure if the bay there would actually be considered a back-arc basin. Maybe just a trapped piece of oceanic crust, something like the Caspian sea. But more on this back-arc basin stuff later.

I blame the difficulty of drawing a polar plate margin in equirectangular format! If only Photoshop had an extension that allowed you to write on a globe... I've sort of redone it, but if it's still not entirely correct it's because drawing in the poles in equirectangular makes me dizzy :p

Hmm, I suppose I hadn't thought too much about the locations of continental crust, especially since the plate directions changed. I've redone the northeastern bit of Molave, moving the islands there and adding some more land so that I can extend the subduction zone. Have I mentioned that drawing geography at the poles in Equirectangular gives me a headache?

I would recommend not subjecting yourself to self-torture like attempting to draw stuff polewards of 60 latitudes in equirectangular . There are better ways to do things, like using GProjector to switch the pole to the equator, then switch it backwards. Like what I've done here (pick Equirectangular Oblique and set the map at "centered -90 latitude"):

Centered on South Pole.PNG

Then I've exported that (without the graticule obviously), drawn you that nice new continent I knew you always wanted there at the South Pole , and uploaded it back to GProjector.

New Continent.PNG

And then reverse the pole shift (again, Equirectangular Oblique and this time "centered on 90 latitude"). Again, export it without the graticule:

Neo-Pangarap.png

The only real limit to doing this are the resolution limits when exporting images with GProjector (theoretical maximum 7500x3750, in practice I haven't been able to export larger than 6800x3400). That said, there was a lot of discussion about switching projections in order to map the poles over in kacey's thread. I haven't tried all the methods suggested there, and it's pretty likely somebody has a better method to do this than GProjector, that doesn't have the same resolution restrictions.

Ok, that makes sense. I'll admit I'm still not entirely sure where the subduction zones should lie since the whole "diversion-subduction zone" is rather... odd... as a whole.

Basically, a "divergent subductive margin" is one that belongs to a plate that has an absolute motion away from the ocean. So, the subduction zone is moving away (or sort of sideways) from the oceanic plate it's subducting. A couple examples:

Eurasia 85 Mya:
85 Mya.png

Eurasia 40 Mya:
40 Mya.png

As you'll note, the Eurasian plate has been kind of moving towards the Indian Ocean for most of its history. Although nowadays I think it has more of an eastwards vector. Anyway, this is why there are a lot of back-arc basins in the Asian margins of the Pacific. The eastern coast of Molave would probably have a similar situation, btw. And a "convergent subductive margin" would look like the following.

Americas 60 Mya:
60 Mya.png

As you can see, the American plates are moving directly towards the subduction zone. Hence, no back-arcs for the Americas, and this is the primary reason why the Asian margins of the Pacific look totally different from the American margins of the Pacific. I hope those pictures explained the concept . In practice, I tend to take things a bit loosely when placing island arcs and back-arc basins, but this is the "science behind it", so to speak.