View Single Post
Old 06-11-2019, 10:57 AM
Quercus is offline
Join Date: Dec 2000
Location: temperate forest
Posts: 7,150
Originally Posted by Chronos View Post
The reason nobody suggested that is because it only works at the equinox. At any other time of year, that line will indeed curve. If I were to do it today, for instance, at noon the shadow would be north of the gnomon, and at sunrise and sunset, the shadow would be south of the gnomon. It's tough to make a straight line do that.
Yes, the tip of a shadow does NOT in general form an east-west line. (I guess it's a good lesson in how 95% of 'how-to's on the internet are just copies of the same, possibly mistaken, thing).

You CAN find E-W from shadows, by connecting the tips of two shadows, but the shadows have to be from exactly the same amount of time before noon and after noon. Conveniently, the shadows will also be the exact same length at those two times.

So, in the morning mark the tip of a shadow from a vertical stick, draw a circle around the stick with the same length as the shadow at that point (so the tip of the shadow is on the circle). Now wait as the shadow shortens (before noon) then lengthens (after noon). When the tip of the shadow hits the circle again (on the east side this time), mark that point. A line through the two points will be exactly east-west. This is much easier and more accurate if you have a length of string (for drawing the circle and for plumbing the stick to be vertical).

I imagine this would be more accurate than finding north by star rotation, but it would require waiting until sunset instead of sunrise to make the call (and of course won't work if it gets cloudy during the day). Not sure if the extra accuracy is really worth waiting 12 hours.