We’ve been predicting eclipses for over 2000 years. Here’s how.
It's amazing what centuries of data can do.
Imagine. You are an ancient human and your reliable and faithful sun suddenly and unexpectedly goes dark. This terrifies you. You think, ‘What if it never comes back? Oh gods, WHAT HAVE WE DONE TO DESER…oh, it’s back. Phew.’ But then, over the years, it keeps happening. You begin to lose trust in the sun’s loyalty and start recording when these events happen. Centuries go by and eventually enough of a pattern has built up that early civilizations are able to predict when these crazy events might occur.
“The idea that it’s not just random is pretty incredible,” says Jonathan Seitz, an associate professor of history at Drexel. “The Mesopotamians figured it out first in part because they had a habit of writing things down. They were doing this because they felt that these things had meaning—they weren’t just random natural phenomena.”
With records stretching back to about 700 BC, Mesopotamians were able to determine the length of a Saros Cycle—the interval between when the Moon, Earth, and Sun line up for an eclipse. A cycle happens once every 18 years, 10 days (11 days on leap years), and eight hours, tracing a shadow on the Earth. That extra eight hours means that the position of the eclipse shifts over time as the Earth rotates.
Though ancient astronomers wouldn’t have been able to monitor all iterations of a Saros cycle (eclipses can occur in the middle of oceans or uninhabited areas), they were able to figure out parts of the timing well enough to know when one might strike. But at this point in history, they just knew the when. Why and how would have to come much later.
Enter the Greeks. For thinkers like Aristotle and others, it wasn’t enough to know that something was happening. It was equally as important to know why it was occurring. “The Greeks became very interested in causation,” Seitz says. The meaning of the eclipse was less important than other factors: “For them, you don’t understand something until you can explain it.”
Greek observations helped figure out how planets move and that the shape of the Earth is a sphere. Without telescopes, they still thought of the moon as a luminous heavenly body, vastly different from our rocky home, but they figured out its relative motion compared to Earth. And even though they thought that the Earth was the center of the Universe, they figured out that an eclipse is the shadow of a new Moon cast by the sun onto the Earth.
Techniques developed by Aristotle and Ptolemy to understand eclipses were in use all the way up until Copernicus and Newton stepped on the scene hundreds of years later.
“That’s not to say that nothing happened in the intervening time,” Seitz adds. People kept building on ancient cultures’ knowledge, accumulating more data, and starting to refine techniques during the Middle Ages. “In the Islamic world in particular, they paid a lot of attention to astronomy and astrology, developed astrolabes to take angles in the heavens, and tried to refine the system,” Seitz says.
Later, thinkers like Tycho Brahe built giant quadrants to make more accurate measurements of the movement of the Sun during eclipses, and some people used techniques to measure the eclipse that we still use today. “They did use pinhole cameras in the medieval period, which lets you measure the magnitude of the eclipse a little better,” Seitz says.
Europe was far from the only place to notice that eclipses were happening. China developed their own eclipse predictions at around the same time as people in the Mediterranean, paralleling the discovery of the patterns of eclipses thanks to their long history of record-keeping. There is evidence that the Mayans also had ways of measuring eclipses, but virtually all their records were brutally destroyed by conquistadors during the European invasion of the Americas.
Despite greater understanding of eclipses, most cultures still saw them as bad omens. Interpretations (slowly) started to change with the advent of telescopes, which revealed the topography of the Moon and allowed eclipse predictions to get much more precise. In fact, in the 1700’s astronomer Edmond Halley made a map of the path of the coming eclipse and published it in the hopes that the general public wouldn’t panic when the Sun briefly disappeared, and that observers might gather more data on how long the eclipse lasted at different locations. The modern era of eclipse observing had finally begun.
“The method we use now is based on something that people came up with in the 19th century” Ernie Wright, a visualization expert at NASA says. The people who started using more modern calculations to predict the eclipse paths were Friedrich Bessel and William Chauvenet.
“Bessel came up with the basic math that we use around 1820, and Chauvenet put it in its modern form in 1855,” Wright says.
Today, we’re able to get even more specific, thanks to our understanding of the shape of the moon. The moon—contrary to every elementary-school drawing you ever labored over—is not in fact, shaped like a banana or a perfect sphere. Like the Earth, it has mountains and valleys that make its shape a little rough around the edges, and that means that it’s shadow is uneven as well.
“19th century methods assume that he moon is smooth, and assume that all observers [are] at sea level,” Wright says. “You have to make those simplifications when you’re doing it on pencil and paper.”
From the late 1940’s until 1963, an astronomer named Charles Burleigh Watts spent countless hours mapping out the variations that appeared on the moon’s surface, focusing on the landforms that appeared on the outer edge of the moon as seen from Earth. His detailed maps helped eclipse predictions get even more precise. Suddenly, the shadow of the eclipse wasn’t an oval, it was a many sided polygon with each angle aligning with a valley on the Moon’s limb.
Then, NASA took it up a notch. The space agency’s Lunar Reconnaissance Orbiter built on Watt’s work, and captured the topography of the moon in detail that would have been impossible to get from photographs of the moon taken on the ground.
Wright took that data about the shape of the moon, the topography of the Earth, and the positions of the sun, Moon and Earth to create an incredibly detailed and accurate accounting of where the eclipse shadow will pass across the United States.
This eclipse is expected to be the most viewed total eclipse in history. And after humanity has spent thousands of years of watching and recording eclipses, there’s still plenty that researchers hope to learn.
“We’ve been talking recently about the fact that there is some uncertainty about the size of the Sun,” Wright says. “It turns out that eclipses are [a] very sensitive method of measuring the radius of the Sun. The Sun’s radius is about 696,000 km. But if you change that radius by 125 km, you change the duration of totality by a full second.”
If a lot of people along the very outermost edge of the predicted path, like those watching in some areas of Saint Louis, or Kansas City, report that they saw plenty of Bailey’s Beads and that it never got fully, one hundred percent dark, then that lets people know that the sun might just be a little bigger than we thought. If it is, that contribution could help future astronomers refine their predictions even more.
As you look up where the path of the eclipse is, and marvel at how modern science figured out exactly when, where, and how a shadow traveling at speeds between 1400 mph and 2500 mph crosses the country, spare a thought to all the generations of people who helped make that possible; from the observers who didn’t know what was happening over hundreds and hundreds of years but still bothered to write down what they saw, to the people who built the modern satellites that made this year’s eclipse maps so accurate. It took all of them to get to where we are today.