NavList:
A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
Re: Refraction
From: Gary LaPook
Date: 2007 Dec 11, 00:30 -0800
From: Gary LaPook
Date: 2007 Dec 11, 00:30 -0800
Here is a link to a site that demonstrates refraction and formulas for calculation. http://www.jgiesen.de/refract/index.html gl On Nov 21, 1:05 pm, Gary LaPookwrote: > Gary writes: > > I am attaching a message that I posted to an Amelia Aerhart list which > goes into many aspects of refraction : > > "Gary wrote: > > Ric has always pushed the idea that Noonan observed sunrise and drew > an LOP based on that observation which he advanced to Howland. Then > Noonan dead reckoned from the first LOP to the advanced LOP and then > started looking for the island. In this scenario Noonan might as well > have dropped his sextant out of the door after sunrise since, > according to Ric, he wouldn't use it again. Ric seems to have jumped > to this conclusion on the basis that the azimuth of the sun at sunrise > was 067� leading to the 157�-337� LOP. He ignores the fact that the > azimuth remained at 067� until 1854 Z, more than an hour after > sunrise. This means that any sight taken during this period would > produce the same 157�-337� LOP. > > Evidence against Noonan observing the sun at sunrise has to do with > refraction. When a navigator talks about refraction he is talking > about the way the light from the sun and other celestial bodies is > bent as it passes through the atmosphere. Because of the density > range of the air all light rays are bent down toward the ground making > objects appear higher in the sky than they actually are. The amount of > this bending depends on how much of the atmosphere the light must > traverse. An object directly over head will have its light bent not at > all and there is no reason to worry about refraction for an altitude > of 90�. As the measured altitude gets lower and lower the effect of > refraction gets greater. The navigation computation tables used by > Noonan (Hydrographic Office (H.O.) 208, Dreisonstok) provided a table > of corrections for the navigator to use to correct for the refraction. > The corrections are given in minutes of arc, 1/60th of a degree. An > error of one minute of arc will cause the resulting LOP to be in error > by one nautical mile. These corrections are always subtracted from the > sextant altitude because the sextant always reads too high. This table > says to use no correction above 70�; 1' between 70� and 36�; 2' down > to 22�; 3' down to 15�; 4' down to 13�; 5' down to 10�; 6' down to 8�; > 7' for 7� and 8' for 6�. That is as low as the table goes. The > equivalent table was also found in the 1937 Nautical Almanac. The > reason these tables go no lower is because at lower altitudes the > refraction becomes much larger and unpredictable so navigators are > trained to not use such low altitudes and the omission of lower > altitudes in the refraction correction tables was meant to discourage > anyone from attempting to use such a low sight. In fact H.O.218, a > more modern set of tables, only allows you to do your computations for > altitudes above 10�. So Noonan couldn't use his correction tables for > any sight below 6�. > > But what if Noonan was such a great navigator that he thought he could > use lower altitudes? Could he just extrapolate from the table that he > had to estimate the correction for sunrise? Well, no. The correction > increases very non-linearly from 8' at 6� to 36' at zero degrees. > However there was a table in The American Practical Navigator (also > referred to as "Bowditch"), H.O. 9., that showed corrections all the > way down to zero so Noonan could have ripped that page out and carried > it with him but there is no proof that he did. > > But that still wouldn't have solved his problem. Since they were > flying at 10,000 feet the visible horizon is actually 1� 37' below > horizontal because he was actually looking down towards it. ( The dip > of the horizon is calculated in minutes of arc as .97 times the square > root of the height in feet. The square root of 10,000 is 100 times .97 > equals 97' or 1� 37'.) This means that at sunrise the actual altitude > measured would also be minus 1� 37' and the refraction table in H.O. 9 > only goes down to zero. Well what if Noonan just used the maximum > correction tabulated for a zero altitude which was 36'? Well the > refraction table found in the modern Air Almanac for sights taken at > 10,000 feet shows the refraction correction for minus 1� 37' is 50', > 14' more than the correction for zero altitude. Noonan couldn't have > known this but he would have known that it was greater than 36' but he > couldn't know how much more. If he applied the 36' correction instead > of the correct 50' correction he would have plotted his LOP 14 NM too > close to Howland. If he took no more sights and just relied on dead > reckoning from there he would have turned 14 NM too soon and could > have missed Howland by being too far to the southwest. > > The bottom line is that Noonan was too smart a navigator to be > ignorant of these problems with refraction. He had plenty of time to > take sights on the sun after it had risen above 6� at about 1815 Z at > Howland. > > BTW, if you watch the sun set over the sea with a clear horizon you > will notice that the shape of the sun changes from round to a > flattened or squished look. This is caused by the rapid changes in > refraction as the sun nears the horizon. The sun is 32' in diameter so > when the bottom is on the horizon the top is 32' higher. The modern > refraction correction table shows the correction for zero is 34.5' > while for 33' it is 28.2'. This means that he bottom of the sun is > refracted up 6.3' more than the top edge is. This makes the sun look > squished since it is still 32' across but only 26' from top to bottom. > Sometimes the sun will take on a lumpy appearance or appear to have > shoulders and this is caused by the erratic changes of refraction that > can take place at low altitudes. When there is a very greatly > increased refraction you can see mirages, objects that you could not > usually see because they are hidden by the horizon. But with extreme > refraction the light is bent so much coming from those objects that it > comes over the horizon and is bent enough to stay near the ground > where you can see it. > > These are the types of problems with low altitude shots, Noonan would > have known about them and would not have attempted low altitude sights." > > gl > > Isonomia wrote: > >Gary, > > >thanks for the pages on refraction. My comments on refraction on the > >horizon were just muddled thinking - of course the sky itself is a > >long way away and is therefore refracted, but the sea which determines > >the line of the horizon is only a few miles away and therefore doesn't > >have as far to suffer. > > >However, as I've often seen a distortion of the sun as it goes beneath > >the horizon I think this is evidence for some kind of affect that > >distorts light at the horizon, which because it happens near the > >horizon must be due to some kind of affect that bends light close to > >the ground and surely this "heat haze" must have a significant impact > >on the position of the horizon? > > >Perhaps a warm sea/cold air is an unlikely observation event (its > >called fog), but a cold sea with warm dry air coming over it must have > >some kind of affect which will alter the observed position of the > >horizon - or does it? > > >Mike > > >On Nov 20, 7:04 pm, Gary LaPook wrote: > > >>Gary LaPook writes: > > >>Attached is an excerpt from Dutton, explaining refraction, and the > >>sextant correction table from the 1999 Nautical Almanac to assist Mike > >>Lenzie. The sun correction table includes refraction and semi diameter > >>for upper and lower limb observations. The included dip table is for > >>correction of height of eye above sea level and can be computed from the > >>formula: dip (in minutes of arc) = .97 times the squaare root of the > >>height of eye (in feet.) > > >> Dutton 1934 refraction.pdf > >>97KDownload > > >> refraction, N.A. 1999.pdf > >>53KDownload --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to NavList@fer3.com To , send email to NavList-@fer3.com -~----------~----~----~----~------~----~------~--~---