The Color of the Land

The land has many colors.  Reds, oranges, tans, whites, and greys tend to predominate in terms of rock colors.  Which colors are most prevalent in a given location varies widely across the world, and even from one place to another within a given state.  Similar land features in different areas can appear vastly different in color and shape.

Capitol Reef National Park, Utah

For example, take the main features of Grand Teton and Capitol Reef National Parks.  The Teton Range of the Rocky Mountains is the main geologic feature of Grand Teton, while the main geologic feature of Capitol Reef is the Waterpocket Fold.  Both are relative narrow (east-west) but long (north-south) ranges formed from uplift.

Teton Range, Grand Teton National Park, Wyoming

When viewed in cross-section, average elevation in both follows a sawtooth shape, being quite asymmetrical; the crest of the Tetons is near the eastern edge of the range, with the crest of the Waterpocket Fold lies along the extreme western edge.  As can be seen in the photographs accompanying this post, they look nothing alike.  Partially, this is because the Tetons have been subjected to extensive glaciation, carving them into the jagged range you see today.

Closer view of the Waterpocket Fold in Capitol Reef National Park

But color is another factor.  The Tetons are largely in shades of gray (and white where the snow lingers) from the bedrock that composes them.  Specifically, the rock of the Tetons is largely gray granite with some black diabase dikes.

The west face of the Waterpocket Fold appears as bands of tan, white, and red from its layers of sedimentary rocks.  The sandstone is generally tan, but may be red if iron was present, while the limestone is the white band; a very pale, almost white, sandstone forms the uppermost layer here.  The Waterpocket Fold is far older, originating with the Circle Cliffs Upwarp during the Laramide orogeny at least 35 million years ago, while the Tetons are only about 9 million years old.  As sedimentary rock overlays bedrock, clearly mere age is not the reason for the difference.

The effects of glaciation can be emulated well enough via erosion algorithms.  Colors, on the other hand, need another source.  Some sort of virtual stratigraphy seems like the answer, but ideally it would involve neither large computational or storage requirements.  Perhaps a fixed, one-dimensional table representing the stratigraphy, with a coarse map generated via Perlin noise indicating for a given point where in the stratigraphy table the local base elevation is to be found (with minor perturbation, perhaps?).  I'm not sure yet, but it is definitely something to think upon.