NavList:

Huub Robroek: Thanks for that link! It's always interesting to see how close we are to Holy Grail: a navigationally accurate celestial smartphone app. The pieces are almost there....

There are two basic principles apparently being used by this system. Some parts of it are described in a possibly misleading fashion.

The most basic problem with any automated celestial system is finding the vertical. Identifying stars and determining their angular positions is child's play. Determining the local vertical? That's tough! The sea horizon, reflecting "artifiiclal horizons" and bubble levels are well-known solutions employed in manual sextants for centuries. From the sound of it, and given one of the references to an article by George Kaplan, this system apparently uses the variation of refraction with altitude to determine the zenith. This technique has some technical challenges. It requires reasonably bright stars at relatively low altitudes. But the principle is straight-forward. It can even be applied to manual sextant observations. If you measure the altitude of a star near the horizon, and then measure the altitude of another star high in the sky on the same azimuth, the difference in observed altitudes compared to the difference in true altitude will yield the difference in refraction which in turn yields the true altitude of the lower star. The rest is easy. Note that this does not work with stars at moderately high altitude. First the refraction differences are smaller, so it's harder to measure. But more importantly the refraction at high altitudes is "center-less". The refraction correction is approximately proportional, at the rate of 1 part in 3000 to angular distance from the zenith or from any arbitrary point within dozens of degrees from the zenith. This means that the behavior of refraction "up high" is useless for finding the local vertical. We need the significantly non-linear variation of the refraction function "down low" near the horizon to make this system work.

The article also discusses using polarization to determine an approximate true azimuth. This is a real phenomenon, and when I have time in my on-site celestial navigation workshops, I often pass out polarizing filters late on the final workshop day, and we try it. The band of sky 90° away from the Sun's position in the sky is significantly polarized. Finding the direction of maximum polarization of the sky is equivalent to finding the azimuth of the Sun. Normally, of course, this is redundant. If you can see the Sun (or the Moon), then you get its azimuth directly, and polarization adds nothing. But sometimes, for example around the time of sunset in the evening, the entire western horizon including the Sun's position may be obscured by case. In that relatively uncommon, but certainly not "rare", case, polarization is a useful surrogate. The article repeats the common urban legend claiming that "Vikings" did this. Can't really fault these folks for that. They're not much interested in history, and they're simply repeating the legend.

Want to try it? Want to use polarized light to determine the Sun's compass direction? Do you carry around polarizing filters or polarizing sunglasses? If you do, great. Take out your sunglasses and rotate them while aimed at the blue sky high overhead. You'll see it. But suppose you don't carry polarizing sunglasses. I bet your phone is in your pocket, and I bet it has a nice shiny finish. Many "shiny" surfaces will work for this purpose, but I find the back side of my phone (no case) works great. Near or after sunset, hold your phone in your hand out directly in front of you. The phone should be level with the horizon (level with your eyes looking at the horizon), and tilted about 45° so that you can see the reflected light of the sky in it. Next turn slowly around, spinning on your heel. You will notice that reflected light from the sky rises and falls in brightness. It's darkest when you are facing two specific directions relative to the Sun's azimuth (towards or away). The phone's surface serves as a polarization filter (see "Brewster Angle" for details). The light of the blue sky is polarized perpendicular to the Sun's azimuth in a fashion almost entirely invisible to our normal vision, but the reflection off a shiny surface detects the polarization. So there's a new navigation function for your GPS-equipped pocket supercomputer! Just don't drop it while you're getting dizzy spinning around. :)

Frank Reed