I assumed the game was trying to normalize some of the anomalous terrain that exists because of how the game world used to be (when the fractal map was added, apparently towns kept their directly adjacent terrain types, even if that made things a bit silly).

So what I'm wondering is, has anyone noticed a pattern of the shifts?  Does it only happen when the prior terrain is a mountain (or do people only notice when mountains suddenly disappear)?  Since the immediately surrounding terrain doesn't seem to affect anything, perhaps the new terrain is just randomly determined from all the applicable terrain types for the biome and elevation?  Or does the alteration not always happen?

Two cities in the last two months changed a large mountain square to normal plains after exodus.

Both large mountains were there long before my cities initially settled on them.

I've seen the newb ring terrain roll over, though I'm not sure exactly when that happens either.  Is it daily?  Or when a newb leaves a square in the ring?  Or when a new newb appears with a new town?

QFT - the system attempts to revert land to what it should have been after the Poussin/fractal terrain, elevation and biome seeding; if the recently evacuated terrain isn't currently what it should be in the eyes of the system.  

The map contains concepts of each square's current terrain type, its underlying terrain type (which is what it was intended to be after the seeding), and its original terrain type (what it was when the map was first created, before the seeding).

Hope that helps.

SC

If so was that intended, if not will it be fixed?

Or is this purely for when our cities leave?

It's only when city spots are vacated, either by abandonment, razing, exodus or, I believe, Tenaril's.

Nothing to worry about, it's always been this way, and we're not changing anything.

SC

Hmm.  Strangely, EF's results seemed to indicate that the terraform would remain, but the town that is on the test spot (-710/844) now is on 5-food.  I wonder if that's because the terraform failed, or someone just teleported a town on top of it.  I wish there was some sort of way to view the history for a specific square.  Oh the stories they would tell...

Yeah, perhaps make it a research... 

...Perhaps some geomancer cutting down a tree to count the year-rings, or something like that   

...oh, yes, and of course this research would be needed to subsequently research terraforming... 

Hmm, something just occurred to me to ask.  I assume the data contained in the datafile_worldmap.txt data dump is the "current" terrain type (since I think I've seen that file change, which the other two terrain types -- underlying and original -- shouldn't); is that correct?

And a follow-up question, assuming that's accurate...I don't suppose it would be worth the effort on your end to ask if you could make the other data sets publicly available?  I can imagine some app developers would be greatly interested in the "underlying" data, to see what the potential of each square is...though perhaps that would spoil the surprise?  (The "original" data I only included so he wouldn't feel left out; who knows though, maybe someone would be interested in looking at historical data.)