[♪ INTRO] Today, we have seismographs all over the world recording the time, location, and magnitude of earthquakes as they happen.

Because knowing how frequent and strong earthquakes tend to be can help people predict the risk of a dangerous quake where they live. But seismographs cannot tell us about earthquakes that happened before they became widely used in the late 1800s.

Fortunately, in some places, ancient buildings have unintentionally preserved seismic data, helping scientists reconstruct earthquakes that happened in the distant past. The field that explores damaged buildings as a way to understand past earthquakes is known as archaeoseismology. And it only really took off in the last three decades.

Part of the reason it is such a new field is because it’s not easy to tell apart earthquake damage from other kinds of deterioration. So it was hard to study scientifically. But more recently, archaeologists have learned to recognize types of structural damage that are often caused by shaking, such as columns that have all fallen over parallel to one another.

Or, if multiple nearby buildings all show the same kind of damage, like tilting or cracking in the same direction, that can also point to an earthquake. And now, computer simulations can help too. Some researchers use 3D laser scanners to create super detailed models of ancient buildings.

And then they can simulate the ground shaking or other phenomena to see what kind of event could cause the damage those rea/ buildings have sustained. Now these methods still are not perfect or foolproof, but they let scientists explore major events that might otherwise be lost to history. For example, in a 2013 study, scientists examined signs of centuries-old earthquakes preserved in ancient Himalayan temples in North India.

They zoomed in on a 600-kilometer region that hasn’t had any major earthquakes recently, but is located along an active fault. So the researchers hoped to figure out what seismic activity there had been in the past, in order to better understand what might happen in the future. To do that, they analyzed the damage to various temples that had been built hundreds of years ago, some as early as the sixth century CE.

One thing they found was signs of an earthquake that happened around the 12th century CE. They figured that out because, within an area of 1000 square kilometers, almost all of the temples that existed during that period had been damaged. Some had signs of vandalism, which might have happened during various invasions.

But others had damage like rotated pillars and shifted stones that was most likely caused by shaking ground. Meanwhile, nearby temples built afterward were still intact, suggesting that they had never been disturbed by a major quake. So basically, if you have structures that were built over a range of years, you can narrow down the timing of an earthquake based on which buildings are damaged and which ones are not.

The authors of the same study also demonstrated a way of using archaeoseismology to narrow down the location of an earthquake. To the north of the 12th-century earthquake, a second earthquake struck in 1803. This one was widely recorded, so the date wasn’t up for debate, but no one knew exactly where it had hit.

So, the researchers looked at where buildings had suffered the most severe damage in the 1803 quake. And they used the locations of these buildings to help narrow down the epicenter to a specific town. In certain cases, archaeoseismology can even offer clues as to how severe a past earthquake was.

Because if an earthquake completely topples a building, that suggests that the earthquake must have been over a certain magnitude to cause that amount of damage. Likewise, if a building is still standing after being hit by an earthquake, it can’t have been that strong. So, by creating computer models of these buildings, researchers can sometimes come up with upper or lower limits for the magnitude of the quake that did the damage.

But in special cases, it’s actually possible to make a more precise estimate. For example, the authors of another 2013 study examined an ancient Roman mausoleum in Turkey. The structure was deformed but still standing, and they suspected the damage had been caused by an earthquake.

There were missing columns, collapsed areas, and sideways-shifted blocks. But the structure had not been toppled. So the authors decided to create a computer model of the mausoleum and run simulations to see what kind of force it would take to recreate the damage they saw.

After testing a bunch of different scenarios, they found that the damage they saw in the mausoleum most closely corresponded with a simulated earthquake of magnitude 6.3 or 6.4. As the field of archaeoseismology grows, it’s being used to shed light on past earthquakes around the world. And it shows that cultural artifacts like buildings can be surprisingly useful as scientific tools.

By using them to explore the earthquakes of the past, we can not only learn about future seismic risks; we can also better understand what impact earthquakes might have had on humans throughout history. Thanks for watching this episode of SciShow, and thanks as always to our patrons for helping to make it happen. You, too, can get involved with our community.

All you have to do is go to patreon.com/scishow. There’s lots of cool stuff you can get there. And most of all, you can help us do the work that we love to do. [♪ OUTRO]