Jari wrote:

"...I'm not sure how correct fact-wise it would be, but wouldn't it be
possible to simulate the shadows of craters based on their diameters.
By assuming a generic rim height and crater profile (appropriately
scaled to diameter, of course) and then have Guide draw the resulting
shadows and sun-side brightening, I think the visual impression would
be stunning."

It could be quite nice, yes. Antonin Rukl's _Atlas of the Moon_ has
some example crater profiles, showing such cases as "peak in the
center", "lava-filled floor", etc. Some mountains and variations in
the rim height could probably be added as well. The result would
probably _look_ quite good, until you compared it to the actual
appearance of the moon.

If we had a detailed grid of elevation data for the moon, computing
shadows would be straightforward (especially if you make them run
east/west, which is almost true on the moon). But the best grid
available (from Clementine data) isn't all that good.

I'm therefore inclined to think that the method of imaging the moon,
night after night, at a variety of colongitudes, is the way to go. Yes,
it'd have to be distributed on CD-ROM. But it would be a tremendously
useful dataset.

-- Bill