Lincoln Ellsworth was not an expert navigator. Bernt Balchen, Umberto Nobile, and Leif Dietrichson all claimed he was incompetent in this area. Nevertheless he did, along with Hollick-Kenyon, manage to bring the Polar Star 2,200 miles (3,500 km) across an unexplored continent to within sixteen miles (25 km) of Little America, which was about the size of a football field and hidden beneath the ice. Had his plane not run out of fuel it is likely that Ellsworth would have flown directly over it. Therefore his flight, from Camp III, high on the Antarctic Plateau, when he knew he was (a) lost and (b) his sextant was set wrong, was pinpoint accurate. How did he and Hollick-Kenyon manage it? Was he guided by some metaphysical force, as polar explorers often claimed they were? Did he manage to navigate the last part of the flight accurately, or was it simply blind luck, or possibly a combination of these factors? To appreciate what a remarkable feat this was, it is necessary to have a basic understanding of the problems of polar navigation in Ellsworth’s time.
Navigation is about angles and imaginary lines on the surface of the Earth. Establishing where they were, in terms of latitude and longitude, along with the direction in which they had to travel, navigators used the techniques and instruments developed for mariners at sea level.
The Earth rotates through 360 degrees every twenty-four hours. This means that every four minutes it rotates one degree of longitude. Establishing which line (or meridian) of longitude they were on, or near, sailors had to know two things accurately: when it was midday where they were and how many hours and minutes had elapsed since it was midday at a place from where their lines of longitude originated. Today this is the Greenwich section of London, and most maps are drawn (as Ellsworth’s were) with lines of longitude expressed as degrees east or west of Greenwich. Knowing Greenwich Mean Time (GMT) for sailors meant carrying one or more accurate timepieces, which were permanently kept at GMT. When ships were in port, or passed other ships, the captains would compare their timepieces for accuracy. By the 1930s radio signals were broadcast exactly on the hour to help navigators check their timepieces. (Ellsworth, in his diary, mentions getting “three pips from Buenos Aires.”)
Establishing exact midday at sea was a practiced art for sailors. It was the time when the sun was highest in the sky. They measured this using a sextant. The angle of the sun’s elevation was measured repeatedly, starting before midday. Elevation and time were noted until the sun stopped rising and began sinking. If they determined, for example, their midday was three hours and four minutes after midday at Greenwich, then they knew they were at 46° West. For sailors, once they had timepieces capable of remaining accurate during the rigors of long sea voyages, the puzzle of longitude was solved.
Nevertheless, longitude held special challenges for the polar aviator. First, in the polar regions, the meridians of longitude converge until they ultimately meet at the North or South Pole. At the equator each degree of longitude is a little more than sixty-nine statute miles (111 km) apart. At the Antarctic Circle (66°33' South) this separation has halved to a little more than thirty miles (55 km). At 80° South, where Ellsworth was flying, it is reduced to twelve miles (19 km). The closer proximity of the meridians makes accurate observations more critical.
Observation of the sun is more difficult in the polar regions. The low angle of the sun on the horizon increases the chance of refraction. The navigator cannot be sure he is looking at the sun and not some refracted image. Wilkins, on his flight across the Arctic in 1928, described the sun “dancing” just above the horizon as a result of refraction. Over the equator the sun arcs from the horizon to high in the sky. Halving the interval between sunrise and sunset gives an approximate midday, and the observer can watch the sun’s arc rise and fall. In the polar summer, however, there is no sunrise or sunset; just one long polar day, during which the sun circles the sky. Determining midday accurately is extremely difficult.
Observations for sailors were aided by the fact they traveled so slowly. Captain Cook’s Resolution needed strong winds to reach five knots (5.8 statute mph). When he circumnavigated Antarctica in 1775–76 he would take hours to go from one meridian of longitude to another. Flying, Lincoln Ellsworth did it in minutes. Landing to take observations was the only sure way for Ellsworth to do so.
Timepieces were also difficult to use in the polar regions. Wilkins wrote that on his flight across the Arctic Ocean his stopwatch ceased to function because it became too cold. He adopted the practice of keeping it under his fur jacket, near his body, to keep it warm.
Lines of latitude run horizontally around the Earth. Unlike lines of longitude, which converge at the North and South Poles, lines of latitude run parallel. They are measured from the equator, which is zero, to the North and South Poles, which are 90° north or south respectively.
To establish latitude the navigator needed to measure his angle north or south of the equator. He did this by measuring the angle of some heavenly body, such as a star or the sun. The polar aviator is flying (at least Ellsworth was) during the long polar summer. There are no stars out. The best astronomical method available, which uses the sun and is relatively simple to employ, is known at the Marcq St. Hilaire method. Using a nautical almanac (a kind of astronomical calendar produced each year) the navigator can determine that the sun is directly above (in its zenith) a certain point of the Earth’s surface. By finding the latitude of that point from the almanac, the navigator can, using a sextant, determine how far he is from it. He measures the angle of the sun with a sextant. He knows the “height” of the sun in the sky, which gives him the length of one side of a right-angled triangle. His sextant has given him the acute angle of that triangle, so using cotangent tables he can determine his distance from the point on the surface of the Earth, directly below the sun. If he knows his meridian of longitude he can establish his latitude.
When the navigator is only able to take sun sights at low elevations, as in the polar latitudes, accuracy becomes paramount. The slightest error in angle read from the sextant is magnified by interpolation of figures from published tables of cotangents. These are least precise at the lowest angles, giving the greatest error for the estimated position on the ground.
This method was devised for sailors who were, obviously, at sea level. The surface of the sea is (ignoring for a moment the curve of the Earth) one side of the right-angled triangle. If the navigator is, for example, at an altitude of 10,000 feet (3,000 m) he must make an allowance, or know the angle from his point of elevation to the point where the sun is directly above sea level (his angle of declination). To measure this angle he needs to know his exact altitude. In an unexplored area, where the heights of mountains has not been established, the only way he can know this accurately is by using a barometer. This measures air pressure, and the higher the navigator, the less air pressure he is experiencing. In the polar regions, the extreme cold and the vagaries of air pressure make barometers unreliable. In short, Ellsworth had no way of accurately knowing his altitude on the Antarctic Plateau, and therefore, even the Marcq St. Hilaire method of establishing latitude could not be relied on.
Knowing where he is on the imaginary grid of lines that encircle the Earth does not enable a navigator to know which way to travel to reach the desired destination. Before the days of radio-direction finding (and, more recently, GPS) there were two ways of doing this. One was using astronomical bodies, such as the sun or the North Star, to establish direction. The other method was the magnetic compass.
The magnetic compass does not, of course, point to the Geographic North Pole, but to North Magnetic Pole. In the northern hemisphere the Magnetic Pole is more than 400 miles (640 km) from the Geographic Pole. In Antarctica, the South Magnetic Pole is some 1,700 miles (2,700 km) from the Geographic South Pole. The position of the South Magnetic Pole was known. But that did not mean Ellsworth could simply use a magnetic compass that would accurately point to the South Magnetic Pole. Being so close to the South Magnetic Pole meant that compasses were highly susceptible to variation. One of the first things polar explorers did when they reached unexplored areas was to chart the magnetic variation. Scott, Mawson, Byrd, and Shackleton all took magneticians to the Antarctic, whose job it was to chart the directions in which the compass needle pointed. Most of the area over which Ellsworth flew was unknown. He knew, once he passed Hearst Land, that he would be flying roughly in the direction of the South Magnetic Pole which, fortunately for him, lay beyond Little America, but how much magnetic variation he would find along the route, he could not be sure.
An additional problem is that magnetic compasses are not attracted to a position on the surface of the Earth. They are actually trying to point to a place somewhere inside the Earth’s core. Near the equator this is not so much of an issue. The difference between Geographic Poles and the Magnetic Poles is known, mapped, and allowed for. Near the Magnetic Poles, however, the compass needle, in an effort to align itself with the lines of the Earth’s magnetic field, actually tries to point downward, into the centre of the Earth. It’s known as magnetic “dip.” The tendency to dip, pulling the compass needle to one side or the other, as it tries to point downward, resulted in the manufacture of the “dipping compass,” which allowed the needle to tilt downward, as well as to the left or right. Ellsworth carried a “dipping compass,” but its accuracy could not be relied on.
His only other means of establishing direction was the sun, which Ellsworth knew at midday would be directly north. He also knew that the sun’s angle changed at an even rate, from which he could approximate north.
In addition to using sun sights and magnetic compasses, Ellsworth had another method of navigating; dead reckoning. Dead reckoning is based on speed, time, and direction. A navigator could set off in a certain direction and, providing he accurately knew his speed and the time he had been flying, and as long as he did not drift off course, he could calculate how far he had traveled, and therefore where he was.
In addition to direction, the other critical factors for Ellsworth to measure for dead reckoning were the plane’s ground speed and whether or not it was drifting to one side or the other.
The Polar Star was fitted with an air-speed indicator on its port wing. This consisted of a simple propeller that was turned by the air pressure as the plane flew. The faster the flight, the faster the propeller turned. Air-speed indicators could not compensate for wind. Head or tailwinds would mean an air-speed indicator could be used as a guide only. A more accurate way of measuring progress was the ground-speed indicator. The navigator looked forward at an angle of forty-five degrees to a point on the ground. That point would be the same distance ahead of the plane as the plane was above the ground. The navigator could start a stopwatch when the sight was taken, and stop it when the plane was directly above the sighted point. The distance traveled, being equal to the altitude, could be compared to the elapsed time, and the ground speed calculated. Ground-speed indicators needed the plane to be traveling over level ground and the plane’s height above that ground to be known accurately. Ellsworth had no way of knowing the height of the mountains below, and therefore no way (even assuming his barometer was correct) of knowing the distance between the Polar Star and the ground. He was also flying over a featureless white landscape. Finding some point ahead from which to take a sighting was almost impossible. Therefore, for his estimates of ground speed, he was forced to rely on the known performance of the Polar Star. Bernt Balchen, the only man to have tested the plane over a long distance, said it cruised at 150 mph. By the time they had reached the Eternity Range, Ellsworth and Hollick-Kenyon knew this was wildly inaccurate. They realized that they would not be making the flight in fourteen hours. Their estimates of time and position, so carefully worked out before the flight, were useless. The discrepancy between the Polar Star’s performance and what Balchen said it was capable of, has never been explained.
Finally, Ellsworth had no way of knowing if wind was causing him to drift to the left or the right. A drift indicator allows the navigator to look forward to a fixed point directly ahead, then notice if he is drifting to one side or the other. With the featureless landscape below him, Ellsworth had few, if any, fixed points.
All these factors conspired against Ellsworth, who was, to begin with, an inexperienced navigator. The flight across Antarctica, by guess or by God, remains an incredible achievement.