NavList:

Since this was discussed extensively just a little while ago, I thought I'd share the experience of some of my students in doing lunars with pixels.   Some of the trials and tribulations may sound familiar to some of you.    

They calibrated their pixel width using stars in Leo, and then did a Sunset - Lunar time difference and did shots of the Moon and Regulus.    One of the bigger issues was pixel bleeding - in order to get a decent image of Regulus, the disk of the Moon became overexposed and quite fuzzy.   They only did as well as 5 degrees in longitude using this, and their conclusion was that navigators "back in the day" had to have considerable patience to do lunars.       

Here are some of the other projects they did

clepsydra

orientation of saxon and norman churches

viking sun stone

map-making using the techniques of Lewis and Clark

Latitude and longitude from sunset/sunrise and meridian passage

Accuracy of backstaves

Nocturnals

Construction of a sun compass

Latitude and longitude from satellite dishes

All of them had to do some statistical analysis of their results.    I was pleasantly surprised by some of the results.   In particular, I was amazed at the accuracy of the sunstone.    They used some optical quality calcite, and used it to find north at sunset.   They were systematically off in their measurements, and were perplexed until I pointed out that the sun was (is) setting south of due west this time of year.   When they put this in, the measurements were spot -on.    As we talked about before, one doesn't know if sunstones were really employed by the vikings.   I'm simply pointing out that they do work, and I was surprised at how well they did.