Artificial Horizon

[nipulsansei]

I got to thinking today, about how artificial horizons have been used since pre-industrial times, to find the horizon when the horizon at sea level couldn't be viewed for whatever reason. Then I got to wondering how this idea came about. Thinking some more, I came to the idea that a person could probably take a clear glass bowl, mark the angle from the bottom up to 90°, then fill the bowl to the 90° line. Even if standing on a slope, this should give a semi-accurate reading... I think.

Am I correct in my thinking, or completely off base?

[Dodgey]

Yes you can use a bowl full of liquid, the problem is that it is affected by the G forces experienced in flight, where as a gyroscopic instrument isn't.

[nipulsansei]

Yes, that would be an issue. Except that I was thinking more along the lines of ground based exploration, or emergency situations if one happened to get lost in the wilderness or somesuch.

For instance: I have a plastic water bottle. Using the angle between the handle of my handy-dandy multi-tool and the blade in relation to the artificial horizon as indicated by the water in the bottle to the relation of a known celestial body, I could - in theory - get a rough estimate of my latitude by making use of the ruler on said multi-tool and the calculator on my phone to calculate the angle.

Having no experience in navigating by latitude, I am unsure if the setup would actually work, however.

[vger]

I think this is exactly how a sextant works. Normally that is used for astro-navigation but that should work if you needed to find the horizon.

[linkxsc]

Well, a bowl with water, or a sphere half full (probably better) graduated on side, along with an external tool (a kind of pointer) would be primitively able to tell your current position... if you were good at knowing the stars.

Add a floaty into it with a magnetized needle, and now youve got yourself a compass also. (Which if you rotated the thing, you can use the existing graduations for compass angle)

[PakledHostage]

Yes, that would be an issue. Except that I was thinking more along the lines of ground based exploration, or emergency situations if one happened to get lost in the wilderness or somesuch.

You'd have a tough time achieving any sort of meaningful accuracy. Remember that one degree of error in latitude is roughly 100 km. You'll have a hard time measuring anything to within a degree. And even if you're lost in the wilderness, you probably know your position within plus/minus 100 km anyway.

By comparison, my sextant's worm gear and vernier allows measurements to within 1/600th of a degree (0.1 minutes). In practice, that type of accuracy isn't practical. Waves on the horizon and atmospheric refraction introduce larger errors than that. Errors due to height of the eye above the ocean surface (dip) are also significant, as are errors in time as small as 4 or 5 seconds.

By carefully judging the location of the true horizon between the waves, compensating for atmospheric refraction and dip using lookup tables and accounting for known drift rates in my chronometer, I can fix a position within a few nautical miles of my GPS position, but it takes careful attention to detail. That's close enough to find the lights, but it also puts the achievements of guys like Frank Worsley into perspective.

If you don't have a good horizon due to fog, haze or some other obstruction, you can use a tray of water as an artificial horizon, but you need to have previously measured how far below the true horizon your pan of water is, for some fixed relative position. It won't work on a moving boat though. You can also use a special scope on your sextant that has a built-in bubble level. Aircraft sextants have these because it is almost impossible to see a sharp horizon from up high.

Edited June 17, 2015 by PakledHostage
Clarified a couple of sentences

[justidutch]

Well, a bowl with water, or a sphere half full (probably better) ...

What if said sphere was half empty?

[YNM]

I'd wait for the day, looking true N/S, getting some rough latitude by the height of Sun at transit... For getting the horizon, looking at water level should be enough. Longitude is done by seeing difference between true noon and watch noon

If hazy: illumination time can help. Not sure of any other way.

[baggers]

Just got a bubble level:

from 1600'.

[PakledHostage]

Longitude is done by seeing difference between true noon and watch noon.

I've tried to fix position by using several shadow length measurements obtained around noon time and plotted against time. The shadow length and time of the resulting curve's lowest point can then be used to calculate latitude and longitude. As I mentioned in my earlier post, the resulting fix isn't very accurate though.

With a sextant on the other hand, each measurement yeilds a line of position. (It is actually a circle on the Earth's surface, but it looks like a line on the scale of most charts for all but the highest objects in the sky.) Take measurements of two different objects and you get two lines of position. Where the two lines of position cross is where you are located. And while the two lines actually cross at two locations, the second will usually be off the chart. It is also possible to shoot the same object (like the Sun) twice to get two lines of position, but you need to wait at least an hour or two between shots. Then if you're moving during that time, you have to offset one or the other of those lines of position based on your ded reckoning. This is called a running fix. Adding a third shot for a third line of position yeilds a triangle near your true position. It is traditional to plot the centre of that triangle as your celestial fix. It is possible to calculate a position within a few miles of your actual position using these methids, and you don't need to wait for a noon day sight to do it. You just need more complex math.

Edited June 18, 2015 by PakledHostage

[YNM]

Hmm... Even the calculated time means comparing to a reference clock, no ? Isn't it basically the same as what I mentioned (mind that noon means sun's azimuth of 180 or 360 degree, and altitude greater than 0) ?

Edited June 18, 2015 by YNM

[PakledHostage]

Hmm... Even the calculated time means comparing to a reference clock, no ?

Yes. I was agreeing with you. Sorry I wasn't more clear. I was just expanding on what you wrote for anyone who might be interested. As you know, one can calculate latitude without knowing the time but an accurate timepiece (chronometer) is vital to all practical methods of determining longitude at sea. This is why the work of John Harrison in developing clocks that would keep accurate time while aboard a moving boat was so important to the British Empire. Prior to Harrison's invention of the chronometer, time was calculated by measuring the distance between the moon and other fixed objects in the sky. This took a lot of math and wasn't easy to do while at sea. Another method was to reference transits of Jupiter's moons using a telescope but, again, it wasn't practical to do this while at sea.