No, we still can’t predict earthquakes (despite what some on Twitter may say)

Elements of criteria can point towards the possibility of an earthquake but, at best, predictions are good guesses

By Troy Farah

Science & Health Editor

Published February 9, 2023 9:00AM (EST)

Smoke billows from Iskenderun Port fire as people walk past collapsed buildings on February 07, 2023 in Iskenderun, Turkey. (Burak Kara/Getty Images)
Smoke billows from Iskenderun Port fire as people walk past collapsed buildings on February 07, 2023 in Iskenderun, Turkey. (Burak Kara/Getty Images)

The world is still reeling from the devastating series of earthquakes that struck Syria and Türkiye this week, injuring more than 68,000 people and killing over 15,000 at the time of this writing. That number is predicted to rise, with some estimates from the United Nations anticipating a death toll as high as 20,000 people, making it one of the deadliest earthquakes this century.

The earthquake occurred at a "triple junction," where the boundaries of three tectonic plates meet: in this case, the African, Anatolian and Arabian plates. As these giant rock formations rub against each other, they release massive amounts of energy. In this case, it struck about 11 miles (18 kilometers) below ground, west of the city of Gaziantep, Türkiye, which is about two hours north of Aleppo, Syria. Nearly 6000 buildings have been destroyed and many victims are unaccounted for.

According to the U.S. Geological Survey (USGS), the quake registered a 7.8 on the Richter scale, a metric used to measure the intensity of an earthquake. Anything 7.0 or greater is capable of causing severe damage over large areas while earthquakes above 8.0 are among the largest earthquakes that can occur.

The first quake, known as the mainshock, was followed by a 7.5 magnitude aftershock about nine hours later, which is rare to have such a strong aftershock. But the region has been pummeled with smaller quakes ever since. And according to some geologists, it could keep on rattling for months and even years.

It may seem like all of this came out of nowhere and it's true that neither Syria or Türkiye had any advance warning of this earthquake. But a tweet posted on Feb 3rd seems to have predicted this exact scenario.

Frank Hoogerbeets, a self-proclaimed earthquake predictor, went somewhat viral this week after posting a map with a series of red circles over the exact area where the earthquake happened three days later. "Sooner or later there will be a ~M 7.5 #earthquake in this region (South-Central Turkey, Jordan, Syria, Lebanon)," he wrote on Twitter on Friday.

While this seems eerily prescient, the USGS has flatly declared that it's not possible to predict earthquakes. In order to accurately predict an earthquake, you must fulfill three criteria: the date and time, the location, and the magnitude. Hoogerbeets got two out of three correct ("sooner or later" is not an exact date or time), though the magnitude was off by 0.3. Close enough? Unfortunately, that still may not be enough to predict an earthquake.

"Statements on social media that an earthquake would happen in the effected region of Turkey were timely given that they were coincidentally made prior to a large earthquake sequence, and the statements were accurate in suggesting that a large earthquake could happen in this region someday because this is a seismically active region with known hazard for large, damaging earthquakes," William Barnhart, assistant coordinator of the USGS Earthquake Hazards Program, told Newsweek. In other words, good guess. "Earthquakes are not a predictable phenomenon. No one can accurately predict the location, magnitude, and timing of an earthquake."

It's also worth noting that Hoogerbeets has predicted earthquakes in the past that never ended up occurring. In 2015, he warned Californians that an 8.8 magnitude would strike, but obviously, it never did. According to a news article from that year, Hoogerbeets did predict an earthquake in Nepal, which killed 8,800 people, two days before it happened. But the article doesn't give much more detail than that. Hoogerbeets did not respond to Salon's request for comment.

Where is Hoogerbeets getting all this data for his predictions? The alignment of the planets, of course. He is associated with the Solar System Geometry Survey, a website and YouTube channel, in which Hoogerbeets reads an earthquake "forecast" much like a weather report. Their theory is that certain planets like Mars or Saturn arranged themselves in a way that gave gravitational tugs on our planet, much like the Moon does to the tides, causing the quakes.

But earthquakes are not related to or predictable like the weather. The Earth's outermost layer, the lithosphere, is made up of plates of rock that are cracked like puzzle pieces. These tectonic plates are constantly shuffling around, which makes it hard to anticipate their next move.

Yet, when they stumble over each other, we definitely feel it. Well, sometimes. Most earthquakes go unnoticed, miles below ground, with about 20,000 per year or 55 per day. The intensity depends on many factors, including complex churnings in the Earth's core. Humans can also cause earthquakes, such as via "fracking," which is blasting fluid into cracks deep in the earth to suck out fossil fuels and oil. Nuclear bombs, volcanoes, hurricanes and more can also influence earthquake activity.

While the exact mechanisms that trigger earthquakes are not fully understood, we can probably rule out other planets causing them. The sun is also probably not implicated in any of this, though that is hotly contested among some planetary experts.

"There is simply no way an alignment of planets can cause an earthquake on Earth. It's literally impossible," astronomer Phil Plait wrote in Slate in 2015. "I've done the math on this before; the maximum combined gravity of all the planets under ideal conditions is still far less than the gravitational influence of the Moon on the Earth, and the Moon at very best has an extremely weak influence on earthquakes."

To disprove this, Hoogerbeets would need to present data showing this influence and probably start correctly guessing a lot more of these events with a lot more precision. Given the sudden attention, it seems unlikely he'll stop making forecasts. Just because Hoogerbeets is likely wrong about the cause of earthquakes being Mars or whatever does not mean earthquake prediction is a worthless endeavor, of course. Clearly such a technology could help prevent a massive amount of suffering if it could be developed. If Hoogerbeets can achieve something like that, then godspeed.

In 1968, writing in the journal Earth-Science Reviews, Tsuneji Rikitake of the Earthquake Research Institute at the University of Tokyo wrote that earthquake prediction technology was imminent. "Present-day development of earthquake prediction research suggests that actual prediction of some class of earthquakes, if not all, may be possible within a period of a few tens of years provided that basic data could steadily be accumulated," Rikitake wrote.

It seems his prediction about this prediction tech was wrong. It's still not easy getting that basic data. There aren't enough sensors deep in the Earth to detect tectonic activity over the timescales needed to create reliable models for such forecasts. This is also part of the reason why we still don't fully understand the physical processes underneath our feet. Research is actively occurring in this area, including using machine learning, but that of course comes with limitations. So maybe solving this problem will take listening to the planet beneath our feet and not folks on social media.

By Troy Farah

Troy Farah is a science and public health journalist whose reporting has appeared in Scientific American, STAT News, Undark, VICE, and others. He co-hosts the drug policy and science podcast Narcotica. His website is and can be found on Twitter at @filth_filler

MORE FROM Troy Farah

Related Topics ------------------------------------------

Earthquakes Fortune Telling Plate Tectonics Prediction Science