Hello from Geoff!
Don't worry about the slightly strange login name - I use it on another board, and it's easier to remember one name than two. I have HandsomeRob to thank for directing me here from there.
I'm currently embroiled in writing a fantasy novel which needs some Maps. In the process of making these Maps I find myself torn between expediency and perfectionism; part of me is happy with something which just shows where places are, but there's a more ambitious part which keeps saying, "you know, it would be nice if you had a proper heightmap for this and could write programs to work out rainfall and so on". Two questions in particular keep coming around a lot; I don't know if this is the correct place to ask them, but I'll repost if it isn't and someone can point me in the right direction.
1. Does anybody know of an algorithm which will take a map of a roughly Europe-sized continent which has coloured blobs to show different levels of heights (up to 200m, 200-500m, 500-1000m, and so on) and turn it into a realistic and believable heightmap?
2. (probably quite hard) If you have a heightmap and know the latitude and longitude of all its points, how easy is it to deduce the corresponding climates?
If there are answers, and they aren't too difficult to implement, I can do the programming myself; I'm just missing the necessary theoretical knowledge. Can anyone help?
Hi, Geoff, and welcome to the Guild. I'm so glad more and more authors such as yourself are finding us and asking questions! I have no help on eithe rof your questions, but I find both interesting and will look forward to reading the response you get.
Make sure to post your maps as well; we'd love to see them and (if ypu wish) help you make them better! Take care,
1. Change the colors to gray scale. Black is the lowest, white is the highest. Blur the whole thing. Drop it in Bryce or other 3D tool as a heightmap. You should be good to go.
2. Google "Earth Biosphere" you'll find exactly what you are looking for.
Oh, and welcome to the Guild! :D
Wow- That is a really good question. Effectively, you are asking how to go from contour lines to a (smoothed) heightfield.
Originally Posted by bricka
There is an example of the blurring method here:
The problem with it is that you will either end up with smooth steps, or loose detail in areas that change rapidly. The worst case example would be for smooth slope on one side of a hill with a drop-off on the other.
Could it really be that easy? I'll give it a try.
Originally Posted by RPMiller
Maybe I should have said "calculate" rather than "deduce". I know enough about meteorology to understand how different climates develop and can work them out in general terms for a map. The problem is getting down to specifics, such as "how far north would a Mediterranean climate extend" and "how high do mountains have to be to block significant amounts of rainfall".
Originally Posted by RPMiller
Hi Geoff! Welcome to the Guild!
Have a look at Fractal Terrains, which generates climate / temp / altitude maps automatically. Here is an example which took about about 10 seconds to create because the map is randomly generated. If you want to create your own continents it'll take longer, but the rainfall / climate conversions are still done by the software. (I have no idea how accurate they are, but there are lots of settings you can play with - axial tilt etc, so I guess they are based on some sort of science).
It is very loose science. ;) The creator, Joe Slayton, has dealt with this very question umpteen times. The basic science calculations are all there, but it should always be considered a suggestion. It does not take things like wind and water currents into consideration since it can't calculate those, and a vast amount of climate conditions are based of wind and water currents. Instead it uses, temperature, altitude and rainfall to calculate the climates which is only half of the calculations for a true biosphere.
Hello Geoff!!! Nice to see you here!!
Regarding climates, you well know how hard it is to calculate what you want. I remember reading a paper about the winds of Argentina (the pampero, the sudestada and another one I can't remember its name...), which gave some figures about speed and humidity change while crossing the Andes. I'll have to go to the library to check it out, though...
However, it would be extremely hard to code...
Idea: How about using a FEA software, like abaqus (which is GNU iirc) or CFD like OpenFOAM (Which is GPL, but I have to say I haven't used -not my field...), although I'm not sure about the last one... I just heard about it from my buddies at uni.
Of course, we end up where we always do: yet another thing to learn to aid our secret vice of conworlding :(
RPMiller hit the nail on the head with his post. Wind and water currents are EXTREMELY important to climate.
Perfect example is the Gulf Stream (travelling from eastern coast of S. America, up over the equator, past the east coast of north america, then crossing the atlantic, and continuing up past the west coast of the united kingdom).
The UK has year-round weather that is more moderate than New York state (I'm here in London, and I'm told that they go whole years without any snow.).
Big deal, you might say--until you compare how far north New York state and the UK are. The Gulf Stream has a HUGE impact on the weather of western Europe.