NavList:

Alex E wrote:

"I tend to accept George's explanation (and most importantly the conclusion that dip uncertainty should INCREASE with the observer's height). The longer path the light ray has to go in the atmosphere between your eye and the horizon leads to larger dip uncertainty."

 

I tend to agree with George here, too, but I don't think this can be settled by theoretical arguments unless someone can produce for us a really good model of the sort of atmospheric conditions that would cause these uncertainties in dip.

 

Theory aside, I've got some evidence to throw at this problem. I used to live in a twelfth floor apartment within sight of Lake Michigan, roughly 120 feet above the lake level. It's a big lake, so there is still a true horizon even at that altitude. There were a number of other buildings closer to the lake partially blocking the view. One in particular, about half a mile away had a line of stonework aligned with the lake horizon as seen from my vantage point. But it was only aligned on average. I would look at it through binoculars (whenever there was a big freighter out on the horizon). Purely for my own entertainment, I calculated how much the horizon was moving up and down. It was frequently a minute of arc out of line with the average and occasionally 2 and 3 minutes of arc. I made no attempt to record weather conditions or anything like that. It was just fun to see the horizon shifting around from one day to the next. So at that height, there IS real variability in observed dip.