Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Riloki]

EDIT: I think I may have discovered a reason as to why this may be the case. Original post below.

I changed all stone layers to be able to contain aquifers for the purpose of an experiment, here's what I've found:
An aquifer layer will form bout 3-5 layers beneath the surface, if the site finder reports an aquifer it'll attempt to place one. Additionally, it makes the entire world an aquifer  And it goes all the way to the bottom of the map.

Therefor based on my findings we can extrapolate the following information about aquifers:

• By default, a certain distance from the surface will be chosen, fairly close to the surface level. I haven't done enough testing to say for sure how far down that is, exactly, on average.
• Once an aquifer is successfully placed, all aquifer-bearing layers below it will also contain aquifers.
• When a stone level or soil level is reached which cannot bear an aquifer, the aquifer ends, but the Z-level immediately under it can still be flooded by that aquifer.
• Depth of soil is not checked to determine where the aquifer may be placed or if it may be placed at all, but an aquifer won't be placed if the map tile doesn't contain any aquifer-bearing layers within (I assume) the number of layers from the surface an aquifer will start.

What this means effectively is that just because the site finder reports an aquifer, sites with 'little soil' will not have one unless the aquifer is only one or two Z-levels below the surface^{Which I don't know is possible.}. Some soil is unlikely, and deep soil is likely to have an aquifer. Exceptions being of course if this site contains conglomerate or sandstone, both of which may contain aquifers.
ORIGINAL POST:

So! I notice that a lot of the biggest most infuraiting (for me) bugs were fixed in the past few versions after an extended leave from DF. Sooo I gen a smaller region, pick a 2x2 spot in a desert (I like glass industries and building in deserts). Now, I know that water has become a lot more necessary for survival, so I made sure to select an area with an aquifer. After all take some wood and stone along and piercing aquifers isn't that major of an endeavour.

Weeellll I load up, there's two layers of black sand, then stone. Great, right? Well... not really. See I started building, figuring aybe there was another biome without an aquifer I didn't notice. Nope, dug shafts all over the map, and finally jsut said 'buck it' and used DF Hack's reveal function.

Not one hint of an aquifer. Not one water bearing tile except around teh caves, and I don't *think* that's an infinite water source. Well I figured 'OK maybe it's just this spot.'

Embarked n an area where I'm sure it only had one biome, and listed an aquifer. There was none. Again. Tried it in deserts, outside of deserts, and in 10 fortresses all advertising an aquifer, only one had even a part of one, and that's odd because it was another solid biome map.


I really have to ask 'what the hay?!' I rely on aquifers for water in a lot of areas. Deserts, tundra, that sort of thing... so I'm wondering 'Is it the fact that i use the mayday download?' 'Is it using DFHack, just loading up with it'? Or is it just buggedi n this version 

[weenog]

I don't know about lying to you about an aquifer, but if the water in the caverns reaches a map edge it seems to be unlimited.  It would have to be, either it's flowing in from the edge, or it's flowing off and a source is replenishing it.  Course drawing water from something exposed to an open map edge comes with its own challenges.

[GenJeFT]

Ive had this problem as well, I deliberately placed fortresses on aquifers there was none. I have yet to see any aquifers at all.

[Riloki]

OK... I finally found an aquifer in a desert so tht CAN happen... but it was in a desert with very deep soil.

What I think may be happening is that DF picks a number of Z levels down to be an aquifer layer.. and it puts an aquifer on all aquifer-bearing tiles. With shallow soil this may be resulting in an aquifer layer existing, but not having any aquifer-bearing tiles to bear it.

I'll test this by making all layers able ot bear aquifers. If I'm correct then one of the same areas that had no aquifer should have one a bit deeper down in stone layers. Will post my findings.

EDIT: This appears to be the case! About 3-5 Z-levels down from the surface I'm finding. ... Incidentally never change all stone into [AQUIFER]. It makes the entire world an aquifer  And it goes all the way to the bottom of the map.

Therefor based on my findings we can extrapolate the following information about aquifers:

• By default, a certain distance from the surface will be chosen, fairly close to the surface level. I haven't done enough testing to say for sure how far down that is, exactly, on average.
• Once an aquifer is successfully placed, all aquifer-bearing layers below it will also contain aquifers.
• When a stone level or soil level is reached which cannot bear an aquifer, the aquifer ends, but the Z-level immediately under it can still be flooded by that aquifer.
• Depth of soil is not checked to determine where the aquifer may be placed or if it may be placed at all, but an aquifer won't be placed if the map tile doesn't contain any aquifer-bearing layers within (I assume) the number of layers from the surface an aquifer will start.

What this means effectively is that just because the site finder reports an aquifer, sites with 'little soil' will not have one unless the aquifer is only one or two Z-levels below the surface^{Which I don't know is possible.}. Some soil is unlikely, and deep soil is likely to have an aquifer. Exceptions being of course if this site contains conglomerate or sandstone, both of which may contain aquifers.

[Callista]

Doesn't the game simulate erosion? Because maybe the aquifer tiles all got eroded away during worldgen?

[Riloki]

Actually I think it's an oversight in the code which can be (provided DF is written in C or a similar language) fixed fairly simply by adding another conditional or two in the aquifer placement code. Or hay, just make aquifers not spawn in locations with no soil or little soil.

Just so I'm not complaining though, the 'workaround' of sorts would be to only embark on areas with deep/very deep soil if you actually want an aquifer. Though actually I may run a few tests to see if collapsing a natural [AQUIFER] material into a  ghost aquifer layer might convert it into an aquifer source. If so I'll record teh steps I take to reproduce it. If not your only choice is to either dig down for cave water or as I said, embark on areas with deeper soil that the aquifer code is likely to catch.