How should land cover be distributed?

[oubiwann]

I had planned on defining a collection of vegetation types, associate them with a normal distribution, and when creating a tile for a given biome pixel, generate subtiles that are representative of that normal distribution.

However, some quick reading on land cover types, and it's clear there is a morphology of how different biomes have their vegetation distributed.

Options for distribution of vegetation types in a biome tile:

• Uniform
• Normal
• Various types of clustering

For the latter, we could use the hexagram30 noise library ...

[oubiwann]

Here's what caused me to open this ticket:

Source: Lotsch, A. (1996). Biome Level Classification of Land Cover At Continental Scales Using Decision Trees [Master's thesis, Boston University]. Boston University Graduate School of Arts And Sciences. http://cybele.bu.edu/download/thdis/alotsch.MA.pdf

[oubiwann]

There's now support in the hexagram30 noise project for generating various random distributions that we can use for this. I think that's sufficient for now.