How humanity measured the shape of Earth
From the stick of Eratosthenes to satellites: two millennia of measurements that pinned down the shape and size of Earth, long before photos from space.
Earth's shape was measured with a stick, not a camera
A common objection to a round Earth is: “we only know it from NASA photos.” That is not true. The shape and size of our planet were determined more than two thousand years before the first photograph from orbit, using shadows, angles and geometry. Many of these measurements can be repeated yourself, and they all lead to the same sphere.
This story is not one brilliant idea but a chain of independent methods that confirmed one another over the centuries. It is that agreement, not any single image, that is the strongest argument.
Around 240 BC: Eratosthenes and two shadows
The Greek scholar Eratosthenes knew that at Syene (today's Aswan), at noon on the summer solstice, the Sun lit the bottom of a deep well, which meant it stood almost exactly overhead. At the same time in Alexandria, about 800 km away, a vertical rod (gnomon) cast a shadow at an angle of 7.2°, one fiftieth of a full circle. The Earth's circumference therefore had to be fifty times the distance between the cities: about 40,000 km.
That is close to today's value (about 40,075 km at the Equator). Keep in mind, though, that we do not know exactly how long the ancient unit of length, the stadion, was, and some of Eratosthenes' errors cancelled each other out. The whole method rests on an assumption the flat model fails: the Sun's rays arrive parallel because the Sun is very far away, and the difference in angles comes from the curvature of the surface. You can repeat the calculation in our Eratosthenes' experiment calculator.
1st century BC: Posidonius and a bright star
Posidonius of Rhodes took a different approach. He noticed that a certain bright star, hidden in most of Greece, barely rose above the horizon at Rhodes. At Alexandria it climbed to one forty-eighth of a full circle. From that difference and the distance between the cities he calculated the Earth's circumference. His result was also close to the real value, although here too some errors cancelled out. What matters more is this: Sun shadows and star positions are two different methods, and they gave a similar picture of a round Earth.
Around the year 1000: Al-Biruni and a single summit
The Persian scholar Al-Biruni devised a method that did not need two distant cities. He first determined the height of a mountain in what is now Pakistan, then measured from its summit the angle by which the horizon dips below level. From the height and that single angle he calculated the Earth's radius. Modern reconstructions put his result close to today's value, but that agreement should be treated with caution: the angle is very small, so a tiny measurement error changes the result a lot, and Al-Biruni did not account for the bending of light in the atmosphere. Our curvature and horizon calculator uses the same geometry: the higher you stand, the farther the horizon reaches.
17th and 18th centuries: meridian arcs and a flattened sphere
As triangulation developed, surveyors began measuring the length of one degree of a meridian arc precisely. In 17th-century France, Jean Picard carried out a measurement that was modern in many respects: he measured his base line with wooden rods, his angles with a telescope, and did his calculations with logarithms. Then a new question arose: is the Earth a perfect sphere?
Newton predicted that a spinning planet should be slightly flattened at the poles. To test this, the French Academy of Sciences sent an expedition to Peru, near the Equator, in 1735, and then a second one to Lapland, near the Arctic Circle. Comparing the length of a degree of arc in the two places confirmed the flattening. The Earth turned out to be not a perfect sphere but a shape close to an ellipsoid, as the physics of a rotating body requires.
Today: the geoid, GPS and satellites
Modern geodesy describes the Earth even more precisely. The geoid is the surface the oceans would take if they could settle freely under the Earth's gravity and the centrifugal force of its rotation. Satellite measurements and navigation work with a mathematical reference ellipsoid, not a flat disc. Photos from space are a confirmation here, not the foundation: they add one more independent line of evidence to those gathered with a rod, the stars and a theodolite.
Why that agreement is decisive
Shadows, stars, the horizon seen from a mountain, meridian arcs, gravity measurements and satellite orbits are independent methods. There is no reason they should agree unless they describe the same real, nearly spherical Earth. A flat model would need a separate explanation for each of them, while a sphere explains them all at once.
Want to check it yourself? Start with the Flat Earth page and its interactive tools, or see how every “flat Earth map” turns out to be an ordinary azimuthal projection of a sphere onto a plane.