Manx – a few responses to your questions.
I believe it has already been mentioned in the thread, but Gary LaPook’s excellent website (https://sites.google.com/site/fredienoonan/home) has quite a lot of information on Noonan’s navigation.
It is fairly common in nautical practice to take a 3-star fix at morning and evening twilight. Stars/planets would be chosen at, ideally, 120deg apart in bearing to give a set of Lines of Position. On a boat or ship, running fix methods are then used to advance the LOPs to obtain a fix Between the twilights, it is usually only single body shots on the Sun, tho occasionally Moon shots can be mixed in. Sun shots can be taken at any time it is visible – a noon shot is often taken to determine local noon.
A nautical sextant needs a well-defined horizon to measure from which implies there must be enough light to see it. An aviation bubble sextant uses the bubble to level the instrument so sights can be taken at any time of day on any object the operator can see. If I remember correctly, the A-7 octant Noonan was likely using has a 2x optic – in clear air, even relatively dim stars can still be seen. In daytime, the bubble is easy to see – in the dark, the bubble needs some sort of illumination. In the early aviation instruments, the illumination was by something quite simple: Radium paint or a battery pack wired to a small rheostat controlling the brightness of a lightbulb – later, when access to aircraft electricity was easy to come by, light bulbs were driven by the aircraft electrical system via cable.
Why does an aviation sextant only have a 2x optic? Partly because an airplane can be a wobbly thing to take shots from – having the wider field of view helps find the body and then keep it in view. Remember, in a periscopic sextant, the only view the navigator has of the outside world is the view thru the instrument – tho the Veeder/Root counter and good preparation helps a great deal.
The air navigator has a couple of challenges taking the shot – the first is, as mentioned, the aircraft is not still. There’s plenty of things moving an aircraft around – vibration from the engines, Dutch roll, people moving about.
The second challenge is that the navigator has to move from window to window while holding a heavy octant. Unless the aircraft has a dome, sometimes aircraft structure (eg: wings, window framing) gets in the way. Sometimes the star is too high to be seen from behind a standard vertical window.
Noonan was a very good navigator. He’d pre-compute his star altitudes and bearings and plan the shot well ahead of time. He also, if I remember correctly, would only take 2-body shots rather than 3.
Taking a 2-body shot is fine. Planning ahead, the navigator would select two stars that are about 90deg different in bearing rather than the 3-body shots and 120deg separation.
Even so, the air navigator of the time would draw a circle around the fix – the size of the circle depends on a number of factors. The error in the fix is why Noonan likely did not aim directly at Howland Island – there is a landfall procedure used at the time where you would deliberately aim left or right of the island destination and then then definitively turn towards the island at the right time.
The good thing about a 2-body and a 3-body shot is that it gives a definite fix. However, once the stars go to bed, the Sun becomes the primary navigation source and because you can only get 1 LOP from the Sun, your fix becomes the Sun line coupled with your estimated ground track. The estimated ground track can be significantly different from your actual track so unless you can get a shot of the Moon at a good angle to the Sun, your daylight celestial fixes are estimates that are only as good as your dead reckoning skills.
To answer your question more directly – yes, 3-body fixes can be shot all night long. It is one reason, I believe, that Noonan planned the flight so that they would arrive at Howland Island near sunrise after a long passage at night – he would be able to take fixes all night (perhaps at 1 hour intervals) so that, in the critical final approach to Howland, he’d minimize the run during daylight when they’d lose the crossing LOP to give a definitive fix.