This question could probably apply even more to the North Pole since I imagine it was reached first. A few months ago, I read the story about Robert Scott’s attempt to be the first to reach the South Pole. He made it only to see that Amundsen had beaten him. The only way I can think of (without using modern technology) to know when you’re close is that the sun would circle around the zenith and maintain a fixed altitude in the sky, but this surely wouldn’t be nearly precise enough to locate the South Pole with enough precision to actually make it within visual range of the camp of the previous person who did. How did they do it and just how precise were they?
Scott knew that Amundsen had beaten him because he found his abandoned camp at 90 degrees South.
He determined his position by a combination of celestial navigation, compass readings, and (I would imagine) triangulation from known reference points, at least at the beginning of his trek.
While not as accurate as GPS, these would be more than adequate to determine your latitude (longitude would be irrelevant at the pole). Also, you could tell just from the length of the day, assuming you had an ephemeris and an accurate timepiece, and the distance you had traveled in a day (the sledges were equipped with odometers, as I recall).
What is amazing is that these early explorers were able to find their way at all given the horrible weather conditions at the poles. I suppose in whiteouts they’d have to rely solely on their magnetic compasses, with allowance for the true position of the magnetic poles.
I don’t see how celestial navigation could work since the sun never sets in the summer. Same for calculating it based on the length of day, which would be either 24 hours or 0 hours most of the time especially as you neared the pole. Odometers on the sledges would no doubt be helpful, but didn’t they abandon them pretty early in the expedition? And finally, the magnetic south pole is so far from the geographic south pole, I don’t see how a compass would be very useful.
I agree that it was amazing that they could find their way given the conditions. I’m going to assume the area around the south pole is very flat and they would have been able to see the camp from maybe a mile away, but I still find it incredible that they could manage to locate it even that precisely about a century ago.
I asked about this a while back:
Also, a related question that I didn’t really get a satisfying answer to:
How are directions expressed when navigating near the north and south poles?
You underestimate earlier generations. Polar explorers used a sextant to measure the elevation of the Sun, from which one’s latitude can be determined. A skilled observer with a good instrument could measure latitude within about a mile, which is within visual range.
Conditions at the Pole during polar summer aren’t that bad, except for the obvious cold. The South Pole is a desert and the Sun shines most of the time.
It works since the Sun never sets in the summer. You can do celestial navigation using any celestial object at all, including the Sun (and in fact, a lot of methods are easier to use with the Sun due to its brightness).
That makes sense though I’m surprised their instruments were that precise. I figure to calculate their latitude to within a mile, they’d have to measure the sun’s altitude to roughly within an arcminute which is a tall order. If I remember correctly, the sun is about two degrees wide so they’d have to be almost perfect in getting the center of the sun aligned with the horizon.
I’ll make a guess on this. They consider Antarctica to have an east side and west side corresponding with the eastern and western hemispheres. Maybe in the same vein, they sometimes consider east and west to be perpendicular with the prime meridian with east being towards the eastern hemisphere and west towards the western hemisphere. If that’s the case, I guess they would make north and south parallel to the prime meridian. I don’t know if this is what they do, but it seems like a logical solution to me.
The sun is a half degree wide, or 30 arcminutes, so to get all the way down to a mile you are talking about centering (or “edging”) to within 1/30 of the apparent diameter of the sun, a bit tight but not OMG how do you do that?!. Relax that to say 5 miles and you are only talking 1/6 the diameter, which IMO would be pretty easy.
I was just reading up on how to use sextants and they use the bottom of the sun, not the center, so that would make it a lot easier. Also, it said professional sextants can measure down to 0.2 arcminutes, which I find simply amazing. The only problem I see now is that unless it’s dead-on summer solstice, the sun is going to be rising or lowering slightly through the day even at the south pole. Did they simply compensate for this by subtracting the expected difference from the observed difference?
I figured they might use the bottom edge rather than the center. As for the sun, it moves through the sky (this is different from the motion of the sky just due to the earths daily rotation) so they would have to keep track of which day it was and what time it was. They would likely look up the position of the sun in a pre calculated table and do some math along with their measurements to get their latitude.
I wonder how accurate their time of day measurement had to be?
Correct. The wiki article (linked above) has a nice little animation of how you line it up.
Fairly accurate, since the Sun can change declination by up to 20 minutes over the course of a day. (More of a problem at the North Pole when they went closer to the equinox.) Not a big deal, however, since accurate timepieces had been available since they were developed for maritime use in the Eighteenth Century.
Okay, so to within an hour gets you to within an arcminute roughly. Time measured to a fraction of an hour gets you to a fraction of an arcminute. Again, sounds like you gotta be careful but its not at a how can they DO that?! level.
I read a book years ago about the race to the north pole (spoiler: Peary was not the first one there). It sounded like they would make more accurate readings at the end of the day or if they were very close to the pole. If I recall correctly, they would take readings at about 3 or 4 points a known distance apart and make some calculations. I think it was start making camp, take a sextant measurement. Walk about a half mile in a given direction, take another measurement. Go the other direction from camp and repeat. I’m not sure of the exact details.
That would make sense in at least 2 ways. First you can get an average. More importantly, it allows you check your accuracy. Lets say you think you can measure and calculate to within a mile. So you take a measurement to get your position. Then you move about a mile further north. Take another measurement, do another calc and get your position. If the further north one shows you are actually further north significantly things are good. If you actually measure/calculate you are significantly further “south” on your more northern position then something is hosed up somewhere.
Upon thinking about it a little further, I think the multiple readings also helped with navigation near the pole. If you can figure out how far each point is from the pole, you should be able to calculate distance and direction to the pole. Assuming that proper trig tables are handy.
It is likely that they used a transit/theodolite instead of a sextant. Easily accurate to within 100m. Any celestial reading creates a “line of position”. You know you’re somewhere on that line. Where two lines-of-position cross (three is better), that’s your position.
Scott used a theodolite; Amundsen and Peary used sextants. Amundsen was all about speed and portability. Peary had both but apparently the theodolite is not as effective near the horizon as the Sun was, nearer the equinox.
As far as I remember, Scott’s book does tell you how he navigated. He used a theodolite rather than a sextant to determine the sun’s elevation. To find the exact position of the pole, he made a number of plots which gave him points around an average position. Eventually, he spotted Amundsen’s tent at the point that Amundsen had determined was the pole or at least within a defined error area. i.e within 1/2 mile. I’m not sure if it was that accurate, but I hope you get the idea.
Modern day directions and positions were normally related to an artificial grid in polar regions. The grid aligns to the Prime Meridian, so it doesn’t have co-ordinates that come to a point at the pole. However, with widespread use of GPS units, points can be defined more easily by lat/long. It’s been a while since I was in Antarctica, so my memory of the specifics has faded, and the grid may have fallen out of common use.
It wouldn’t have to be very accurate at all if they were taking multiple measurements throughout the day as was said by someone earlier. They’d only need to know the rate at which the sun was rising or falling at the pole, not at what altitude it should be, and I imagine they could get by just knowing what the date was. Thinking about it, it would probably be a fairly easy calculation really and they wouldn’t need a book.
For instance, if they took measurements of the sun’s altitude 12 hours apart from a given location obviously in opposite directions and the second measurement showed the sun had dropped two arcminutes but it was expected to drop three at that time of year, it was effectively higher in that direction and they need to go roughly towards the direction they measured first.
Of course they’d need to take more than two measurements throughout the day to get an accurate direction and I’m not saying this is the method they used, it was simply the method mentioned earlier.
That’s very similar to the guess I made. Did they have names for the directions along this grid?
As it aligns with the prime meridian, the direction pointing “up” along 0 degrees is grid north. Which means at Scott Base/McMurdo Station (around 166 degrees east), grid north points more or less to the south.
To specify a position, the map is defined in grid squares, which are named in a two letter, two number fashion (I think), covering the continent. Within a square, use 60 by 60 grid. I tried looking on line, but I haven’t found one of the maps yet. I did find an overview type page which shows the area covered by the grid. Polar Stereographic Data