NavList:

Jeff, you wrote:

"Here's an image showing the difference I'm talking
about: http://www.jeffschroeder.us/celnav/snip1.gif"

 

Thanks for posting the side-by-side comparison. I was curious to see it.

 

It's interesting to consider how small a difference this change actually represents. The critical altitude shifts by three minutes of arc from 43d 59' (2003) to 43d 56' (2005), which seems like rather a lot at first glance. Of course, this number is the altitude where the total correction rounds to 15.2 minutes instead of 15.3 minutes. That is, it's the altitude where the total correction would be almost exactly 15.25 minutes. I wanted to see how much of a change in the refraction calculation could move that point by three minutes. The refraction is changing very slowly, so a small change in the calculation should be able to shift that critical value quite a bit. I put together a little spreadsheet using simplified expressions for the various factors (total=SD-k*tan(z)+HP*cos(h) which should still cover the behavior properly) and found that this 3 minute shift in the entry from 43 59 to 43 56 would result from a one-tenth of a second of arc change in the refraction constant. Yes, one-tenth of a SECOND of arc. That's totally insignificant for navigatin, and it makes me wonder why anyone would bother re-calculating the table. A small error that is literally more than a hundred times larger than this refraction difference results from the Sun correction table's assumption that the semi-diameter of the Sun jumps by a couple of tenths of a minute of arc on April 1 and October 1.