NavList:

A Greg has already said, yes, 9° is a more reasonable ending for useful celestial navigation, and you can start much earlier. The idea that practical twilight ends when the Sun's true altitude is 12° below the horizon seems to have emerged with the creation of the term "nautical twilight" (when? sometime in the 1920s?). I'm convinced that -12° was selected for the most trivial reason: to split the difference between the limits of civil and astronomical twilight which were already well-establsihed at -6° and -18° respectively. It's an arbitrary number.

I wouldn't say that the solution is to replace an arbitrary -12° with an (also) arbitrary -9°, let along the 8°51' supposedly measured by German mariners. There's great variety in practical instances. At the limit of practical nautical twilight, direction makes a huge difference. In most parts of the world, you can save the stars in the west after sunset until the end of a round of sights since the horizon is visible much longer. The eastern horizon after sunset can be difficult when the Sun is as little as 7 or 8° below the horizon. In the early morning, we can shoot stars in the east well before any reasonable horizon appears on the opposite side of the sky.

Also, celestial navigators in high latitudes in the right season can frequently see a horizon all night long. Same Sun depression rules fundamentally. Just worth remembering that the Sun can be due north and be high enough to illuminate the horizon above it due north (or due south in the right season in the southern hemisphere).

The rise of apps has led many software-driven navigators to try to craft rules that could feed a practical navigator the right suggestions by algorithm. So far I have not seen anything really effective, and given the large number of sources of variability of the stars and the horizon, I tend to think projects like these are a bit hopeless. There is advice worth knowing, but in the event a navigator has to make selections based on the sky at the time.

Frank Reed