Page 1 of 13 12311 ... LastLast
Results 1 to 10 of 127

Thread: How to get your rivers in the right place

  1. #1
    Software Dev/Rep Redrobes's Avatar
    Join Date
    Dec 2007
    Location
    England
    Posts
    4,807
    Blog Entries
    4

    Post How to get your rivers in the right place

    We have had so much talk on this subject and so many maps fail this point that there is a River Police badge now for officers of the CGuild to catch miscreants in the process of making a map with duff rivers.

    We have stated many times that water flows down hill and others, especially Waldronate, have given a list of points that must be true for water in a non magical and natural terrain and yet we still have those maps coming in.

    So this tut is not so much about what we should be doing with our rivers but exactly how to marry the real world with the pen in your hand.

    So first lets state a few really basic known properties about water in a non magical terrain.

    1. Water flows downhill and it does so in the direction of steepest decent.
    2. Rivers originally start with water which has fallen from rain.
    3. Rivers ultimately end in the sea or in rare cases may evaporate into nothing.

    We will add more shortly but lets dwell on these for a second.

    Point 1: Rivers flow downhill.

    If you don't know what is high ground and low ground about your terrain then you cant get your rivers traveling in the right direction. Trying to make terrain heights based on already specified river patterns is hard work. So determine your terrain height FIRST.

    The statement also implies that two separate bits of water will always follow the same path. This is generally true. So when two rivers meet they BOTH then travel in the SAME direction. I.e. rivers do not spontaneously fork or split into two and go different ways around an obstacle. Only in a situation where one path cannot take the combined flow from the source does the water then split into different routes. So rivers always join up and do not split up.

    Since rivers travel in the direction of greatest decent you cannot have rivers on the top of a hill or along any ridge. In fact when a rain drop falls it goes into exactly one 'Catchment Area' and these areas are usually separated by hills and ridges. A drop of rain just one side of a ridge falls into one catchment area and another just the other side of the ridge perhaps a few feet away go into a different catchment area. Those catchment areas collect the water into streams and rivers and form a 'Drainage basin' and each basin will eventually have exactly one river to the sea. So find the high points in your terrain and divide up your terrain into catchment areas based on them.

    If your river enters a zone that in all directions means that it now has to go uphill then you are in a basin. Only in a basin will a lake possibly form. Lakes are a much scarcer geographical item than most fantasy maps have them. When we have a lake it can do one or more of the following :-

    1. Fill up until it overflows out with another river.
    2. Seep into the ground (i.e. overflow into the ground)
    3. Completely evaporates. This is also very rare indeed.

    If the lake overflows to another river then it is true that it flows out with exactly one river. It does not fork rivers out from a lake unless there is a temporary swell in rainfall, causing flooding. This is true because if there were two exits from a lake then the level would fall until there were just the lower one to exit the water. If you had two exits the same height then one would erode and the other would silt up. Which one is down to the chaoctic nature of the universe but in geological terms it would happen very quickly.

    So you can have multiple rivers entering a lake but only one exit river. In effect a lake is a bit like a very fat bit of river.


    A bit more complexity.
    ================

    Rain or a river can hit an area of rock that has big cracks in it or is very permeable. In either case the water can go underground. From here it can gush out of another big crack or cave mouth. The cave mouth can be above or below the level of the surface of a lake. This is a spring.

    When rain permeates through pourus ground then it will hit a water table. This is a layer of saturated ground and the water which makes it up travels towards a spring very slowly. The 'surface' of the water table might be horizontal underground or it might vary in height undulating. Where the surface ground terrain height drops away to a level below that of the water table thats where springs will occur and water will drain from the water table feeding the spring. Spring water which has permeated for a very long time in the water table is the stuff thats very pure and often safe to drink. So springs occur part way up a hill side and they are always at a very similar level - i.e. they follow the contour of the terrain. You have to dig a well to a depth to reach past the water table before the well fills with water.

    So a bit of terrain comprising of a hill with a cliff on one side cant have wells on the hill. The rain falling on the hill goes into the ground or runs away. That going into the ground will fall out of a cave somewhere on, or at the bottom of the cliff.


    Exceptions.
    =========

    Not many. The main one is that rivers erode terrain and pick up sediment. It can drop this sediment which then prevents the water from flowing in that direction so it much change. Thats called meandering. When the river has passed over a rapids in comparatively soft terrain it will be loaded with sediment. When it then flattens out and slows down it will dump off much of that sediment causing lots of meandering. In those situations the river can 'braid'. It splits and forks into many small rivers all flowing in approximately the same direction. In all but the most rare cases will that river rejoin and come back into one channel again tho.

    joão paulo has pointed out that the largest island where a river has forked around it is a mere 20 square kilometers.
    http://www.cartographersguild.com/sh...28&postcount=3
    This river is going through its very temporary meanderings to find one course through this region. One day soon one side will dry up. In any case 20 square km is very small.

    When a river carries a lot of sediment it can dump it off right next to the sea. For the same reasons the river will braid but instead of rejoining, it hits the sea first. In this case it can form a delta where it spreads out into the sea. These deltas are usually small in comparison to the length of the river. The Nile delta is about 100 miles north-south but then the Nile is 4000 miles long. At the very least a river should have to pick up enough sediment so that its well loaded by the time the delta STARTS.

  2. #2
    Software Dev/Rep Redrobes's Avatar
    Join Date
    Dec 2007
    Location
    England
    Posts
    4,807
    Blog Entries
    4

    Post

    Ok so far this has been said before so I will show an example of what this means and hope that the example will help people get their rivers is better shape.

    I want to make a terrain. To start with we need the terrain height so I will draw on some colored bands to show where the high areas will be.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	ContoursColor.png 
Views:	588 
Size:	10.3 KB 
ID:	8836  

  3. #3
    Software Dev/Rep Redrobes's Avatar
    Join Date
    Dec 2007
    Location
    England
    Posts
    4,807
    Blog Entries
    4

    Post

    If I mark on this map where the catchment areas will be and the direction of water within them then you will see that it will flow towards the middle where there will be some uncertainty about what it would do next.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	ContoursFlow.png 
Views:	458 
Size:	58.7 KB 
ID:	8837  

  4. #4
    Software Dev/Rep Redrobes's Avatar
    Join Date
    Dec 2007
    Location
    England
    Posts
    4,807
    Blog Entries
    4

    Post

    In my example there are two exits off of the map where water could reach the sea. I am going to show later that water will chose just one of these. If you put down your rivers so that they followed the blue lines then that would be good. Here is a diagram with the blue lines perpendicular to the contour lines. Our rivers should be on these lines somewhere. What were expecting is that a lake with rivers should form in the thick blue section. If one side of the map had an inflow of water then it would run into the lake and out of the other side. If there was no inflow then the lake would drain to one side only because you can only have one exit. What happens to the river bottom right depends on whats past the bottom right side but we should get something here because all the water within that catchment area has to go somewhere.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	ContoursPerp.png 
Views:	423 
Size:	65.4 KB 
ID:	8838  

  5. #5
    Software Dev/Rep Redrobes's Avatar
    Join Date
    Dec 2007
    Location
    England
    Posts
    4,807
    Blog Entries
    4

    Post

    So providing you come up with something that fits the bill then it would be alright.

    I have run a simulation and got it to predict where the rivers would be based on some rainfall. I set it up so that either side of the map was an exit for water. The result is below. The middle part of the map started as a lake and then drained to one side. The middle of the map is an area of confusion so almost anything could have happened and it would have been reasonable.

    What was fixed tho is that all the rivers follow the blue lines meaning that they are perpendicular to the contour lines and thus running downhill as fast as it can. Also no rivers crossed the pink lines of the catchment area borders because those lines determine which way water flows. Thats how we can predict where the rivers should be on a map without having to resort to a simulation.

    I am attaching a movie in MPG format to show the simulation running. I hope this will show whats going on and why it was fairly easy to predict where the rivers would have been from the height terrain. Though a lake forms in the middle it is unsustainable and one side won out - in this case the right hand side.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	Final.png 
Views:	715 
Size:	32.8 KB 
ID:	8839  
    Attached Files Attached Files
    Last edited by Redrobes; 01-01-2009 at 09:29 PM.

  6. #6

    Default

    Very interesting stuff - I had no idea about most of this, but it all follows and makes sense when one thinks about it.

    I have a couple of questions:

    - It looks to me as if the lake would drain to the right in your example anyway, because it looks like the terrain rises up to the yellows/oranges just off the the left of the map. So the fact that it does do that makes sense to me. I'm not sure what the river in the bottom-left quadrant is doing though, it seems to fill into the lake and then the lake retreats from it into the top half of your map and then drains to the right? So where is the river water going from the bottom left while it's doing that, is it just seeping into the ground? (also, what software are you using to do the simulation?)

    - One thing I've found curious in the real world is northern canada - if you go look at it on google maps ( http://maps.google.com/ ) you'll see there that a huge swathe of the country around Hudson Bay, is covered with small lakes (you can see this in the map mode really clearly). Where do all those come from? It looks to me like the water table is above the land surface here, is that right? Or is something else going on? (e.g. permafrost in the ground is having an effect?). Actually looking at it, northern Ontario nearest Hudson Bay is actually lacking these small lakes (there are several rivers going to HB instead), and the same goes for southwest of HB, but to the east and west and northwest of HB we're back to lots of lakes. I suspect the geology has an effect here (is the precambrian shield that makes up a lot of canada impermeable, maybe?). That may be one thing to explain here too - the effect of permeable vs impermeable rock that the water is going over.

    Thanks for doing this anyway, really interesting stuff!
    Last edited by EDG; 01-01-2009 at 10:47 PM.

  7. #7
    Professional Artist Nomadic's Avatar
    Join Date
    Aug 2008
    Location
    The great northwest
    Posts
    353

    Post

    Quote Originally Posted by Redrobes View Post
    2. Rivers originally start with water which has fallen from rain.
    A river can also be the result of melt water from mountain snowpacks. In fact that is where most of the water comes from where I live (central oregon). When you have a river like this you get high and low points. The highest points usually come during spring and early summer when the snowpack starts melting. Late summer and into fall it tapers off, getting to its lowest in the winter. Winter replenishes the snowpack and the cycle starts over again.

    Good look at how rivers form though, very helpful (have some rep).

  8. #8
    Guild Journeyer
    Join Date
    May 2008
    Location
    PEI, Canada
    Posts
    213

    Default

    Being Canadian I have to question your view on lakes being more rare than they are usually presented in Fantasy. Honestly most maps I've seen tend to lack lakes in a number I would expect when placed in a northern region that would be subjected to glacier action.

  9. #9
    Guild Member Asharad's Avatar
    Join Date
    Dec 2008
    Location
    USA
    Posts
    80

    Post

    Quote Originally Posted by Talroth View Post
    Being Canadian I have to question your view on lakes being more rare than they are usually presented in Fantasy. Honestly most maps I've seen tend to lack lakes in a number I would expect when placed in a northern region that would be subjected to glacier action.
    Yep, some areas are covered in lakes. If you've flown over southern Louisiana and SE Texas, you know what I mean. Here's an example from google maps.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	SNAG-0001.jpg 
Views:	1357 
Size:	63.6 KB 
ID:	8846  

  10. #10
    Software Dev/Rep Redrobes's Avatar
    Join Date
    Dec 2007
    Location
    England
    Posts
    4,807
    Blog Entries
    4

    Post

    EDG:
    The terrain might raise up to the left but there was a single flat base level which went across the middle of the map to either side. It could have gone either way. It would have been reasonable to have a river going east to west or west to east.

    During the sim if the water height is not very much then it does not show so if the water spreads out then it can vanish from the view. I set it not to permeate or evaporate and its running on my GeoTerSys app. There is in fact rain and water all over the map just not very much of it. The app is not ideal either. Its not without problems and differences with the real world. It is objective however, so it can come up with an answer which lacks my bias.

    Talroth, Asharad:
    Perhaps there are more lakes than I thought. It was a personal observation not really steeped in fact. Our world has some crazy places on it which defies all sensible mapping. Heres a good one...

    http://maps.google.co.uk/maps?hl=en&...37708&t=h&z=11

    It seems that there is a band across the world from Canada, Norway, Finland, and right across Russia in a small set of latitudes which do have a tremendous amount of lakes. Below that line they get a lot less numerous. But best ignore that statement - some places do have a lot of lakes. I dont know why this is or whether it has anything to do with water tables. I would doubt it. I wonder because of the latitudes that it might be because for some part of the year lots of water is locked up as snow and ice and then later it all thaws into lakes and then refreezes again. Maybe that climate precludes long rivers because theres no continuous steady water flow.

    Nomadic:
    Sure, should have said rain and snow. It was basically that water has to gain the energy to rise up in height before losing it all as it falls back to the sea.

    I am not much of a geographer but have been wrestling with some of these things trying to model them in water flow to try and get the results looking natural. We could do with more expertise with geographers and geologists on this site.
    Last edited by Redrobes; 01-02-2009 at 12:39 PM.

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •