I am only looking at which sides of the buildings are in shadow.
AM then W or NW side of building in shadow.
PM then E or NE side of building in shadow.
for example the large building at bottom right of picture.
I recon that it is pretty much oriented so walls are along cardinal directions.
The visible "right" side is North. It's logically in shadow.
The visible "left" side is East. It's in the sun. Thus AM.
Not too much detail on shadow length but best guess is mid-morning.
I believe there are programs that will calculate the unknowns using known values (like height of the building) in a picture.
I'd like to play with such a program.
> How do you work out what direction the shadows are? You must know from what direction the picture was taken? I can understand working out that it was either earlier morning or later evening, based on the length of the shadows....
Well, if you draw a line parallel with the line that a vertical element makes on nearly level ground, that's the azimuth component of the Sun at the time. In that photo, I'd say the Sun's azimuth was approximately 135°.
If you draw a line (in perspective) parallel with a ray connecting some point on a shadow with the same point that cast the shadow, you have the altitude component of the Sun when it is projected onto a vertical plane oriented to the Sun's aziumth. In that photo, it looks as if the Sun's altitude was approximately 62°.
Knowing the latitude and longitude of Marfa, Texas, one could even estimate the time of day and day of year that photo was taken. My estimate would around either April 10 or August 30.
