NavList:

The discussion on dip short and refraction models solidifed some of my own questioning thoughts on the whole subject of refraction in general.  

The second thing that had got me thinking was glancing at the sun rising at Block Island this week and blissfully thinking what a wonderful improvised sextant it would be to just record the time the lower limb just was touching the horizon. Voila...no sextant needed.....except as I understand it.....for low altitude refraction...

I understand from poking around that Shackleton and others have verified that low altitude refraction can change low altitude perceptions from 2 to 4 degrees or more.

I also understand mirages to be a low altitude refraction issues that can be either inferior or superior because low altitude refraction can in fact act in either direction under certain conditions.

So this leads me to the following questioning thoughts...

If it helps, I am splitting how refraction impacts two things....

 a)  light rays coming to me from the astronomical body

 b) near terrestrial perceptions of position and elevation of the horizon.

So:

1. Atmospheric floating objects (reflections, apparent pools, floating camels, ships, etc...)  and such phenomena imply that refraction effects are significant over visual range distances on terrestrial relationships. Does this mean that refraction alters my perception of the true horizon line when I take my sights?, and if so how do I know other than visual clues "the atmosphere is wierd today".  Even with a high altitude sight, if my horizon perception is altered, I am still in error. What actual potential deviation error exists on horizon line perception?

 2. Now for the astronomical body itself....It seems like other than readily available low altitude tables and T & P corrections, low altitude sights such as the improvised sextant described would be very subject to our inability to predict the exact conditions throughout the lights pathway from the body. Simply put it may be passing through varying  conditions from night time cooling, stratification, inversions, weather fronts, etc......so this goes out other than ..I got nothing better?

I see handout or test problems sometimes being worked to Gnat's eyeball accuracy....which does prove I can do equations and find the theoretical numbers....but also critically important to me to learn,  if I claim to be a "practical" navigator is : "how wrong might this position input data be". I jokingly tell people, you really don't need to know exactly where you are, as long as you are very sure of where you are not.(I probobly stole that, but can't remember from who, so attribution is deserved, but unknown).

As a side note: Seemingly harmless, in real operations, trying to be more accurate than the inputs really support is a sickness to fight .....It is  ultimately a distraction which has an impact especially operating alone:

1. In some cases it increases my chances of dumb math errors.

2. It sometimes gives me more faith in a single input than is warranted by its true data reliability.

3. It distracts me from developing alternate inputs that may serve as more valuable cross checks.

MC