We may finally know why the Arecibo Telescope collapsed

The Arecibo Telescope in Puerto Rico defined a generation of radio astronomy, It was commissioned in 1963. For over 50 years, until 2016, it was the world’s largest radio telescope. It tragically collapsed in 2020, and decommissioning work is ongoing.

But what happened exactly? Why did the observatory collapse, when by all accounts, it was structurally sound? Why did the wires snap when they weren’t bearing their maximum load?

A new report from the National Academies, National Science Foundation, and the University of Central Florida focuses on the Arecibo Telescope collapse — specifically, the failure of the zinc-spelter sockets — and gives us new information on what the root cause might have been.

Why is Arecibo important?

You may be wondering why I’m talking about a telescope that collapsed four years ago, but Arecibo was— and still is—iconic. Arecibo has a unique place in pop culture, but it’s also made many important contributions to science. NASA’s radar transmitter at Arecibo meant that it was important for tracking asteroids that could threaten the Earth.

Scientists found the first exoplanets, or planets outside our solar system, thanks to Arecibo. The funny part is scientists were able to make the observation because the telescope was undergoing repairs, so it was looking out at one fixed part of the sky. It was able to pick up small fluctuations in radio bursts from a pulsar over a long period of time, which scientists decoded as small tugs, or wobbles, on the star resulting from planets orbiting it.

For many of us, Arecibo represented something bigger: the small hope that someone is out there, that there’s intelligent life somewhere in the universe looking for us. It’s a belief in something greater, in the cosmos itself.

For me, Arecibo captured my imagination thanks to Carl Sagan’s book Contact. When I saw the film starring Jodie Foster, I was immediately taken in by the romanticism of this place in the jungle where people were working towards something better. (You may also have seen the observatory in the James Bond film GoldenEye, and it is the subject of one of 116 photos on Voyager’s Golden Record.)

Hurricane Maria hit Puerto Rico in 2017 and did serious damage to the island, but we thought Arecibo was spared. It was 2020 that was devastating for the telescope, and honestly, for anyone connected to the observatory (whether they’d worked there, done research with the observatory, or just had an emotional connection like my own), it felt personal.

On August 7, 2020, one of the observatory’s cables broke. Another followed in November, and on November 19, the NSF announced that the telescope would be decommissioned.

Then, on December 1, 2020, the instrument platform collapsed into the dish. Thankfully no one was harmed. But it had been a hard year, we’d lost so much already. And now Arecibo was gone. I was devastated. I still am.

Time has passed, but questions remain. Why did Arecibo collapse? It shouldn’t have.

How Arecibo was designed to work

To understand what’s going on here, let’s talk a little bit about how Arecibo was designed.

The primary reflector dish was a spherical cap made of over 38,000 aluminum panels, suspended just above the ground thanks to steel cables over a natural sinkhole.

It was 1000 feet or 305 m in diameter, built in 1963, and upgraded several times over the decades, most significantly in 1997.

Because the primary dish was so big, unlike other radio telescopes, it didn’t move. Instead, there was a focal structure suspended 150 m or 900 feet above it, weighing 900 short tons/803 long tonnes. It was also where all the receiving equipment was stored. Scientists would “point” the telescope by moving receivers on the suspended platform. The Gregorian dome, installed in 1997, helped focus radio waves onto the receivers.

This suspended structure was attached to three concrete towers, confusingly named Towers 4, 8, and 12, with steel cables. When Arecibo was originally built in 1963, there were four cables per tower, making a total of 12 main cables. The concrete towers were then tied to the ground with 5 cables each called backstays, so that’s a total of 15 backstays, from each tower top. The suspended platform was also tied directly to the ground with six tiedowns.

Now as I mentioned in 1997, the Gregorian dome was added, which significantly increased the weight of the platform — it was around 40% heavier. More supports were added — more auxiliary cables, with two additional backstay cables per tower.

At this point there were over 4 miles of steel cables supporting Arecibo. And when I say cable, I don’t mean a simple cable. These were made up of 126 to 216 galvanized steel wires in concentric layers woven into a single strand. The number of wires depended on the cable type.

All of these cables were anchored into zinc-filled spelter sockets, which will become important later.

The collapse of Arecibo

Okay, so. Let’s talk about the collapse.

Here’s the thing — even though it felt sudden, it didn’t actually happen at all once.

Hurricane Maria slammed into Puerto Rico as a Category 4 storm on September 20, 2017, and that was the beginning of the end for Arecibo. We don’t actually know how much damage Maria did because of “sparse inspection documentation,” according to the NSF report, but it seemed to be minimal. I’ll talk more about that in a minute.

But then, almost three years later, on August 10, 2020, an auxiliary cable attached to Tower 4, pulled out of its socket. The socket failed. Because of the tension, when it pulled out of the socket, it hit the Gregorian dome and then crashed onto the primary reflector dish. The main cables held, but their weight load increased due to the missing auxiliary cable.

The damage was bad, as you can see above. The NSF assessed the situation and put together a repair plan, but they were hampered by the fact that there was no obvious cause for the socket failure. Parts were ordered and repair work was scheduled to begin on November 9.

Then, on November 6, one of the main cables attached to Tower 4 failed, also pulling out of its socket. At this point, the National Science Foundation determined that it wasn’t safe to conduct repairs on the observatory, leading to the November 19 announcement that Arecibo would be decommissioned. The observatory would be demolished rather than repaired because they could not assure the safety of engineers and repair workers.

On December 1, one more of the three main cables attached to Tower 4 failed, once again due to its socket. The cables that were left simply could not handle the weight of the suspended platform, and it collapsed into the reflector dish.

I’ve created basic GIFs of two videos from the National Science Foundation below, but it’s worth checking out the full video.

This first one shows the top of Tower 4, as taken from a drone that was designed to monitor these cables to ensure the health of Arecibo’s support structures. You can already see some fraying wires, and you can see some paint chipping on the middle cables due to wire breaks.

Then, the collapse. It all happens really fast.

Okay, here’s another view from the platform. This was taken from the operations building at Arecibo. That’s Tower 4 in the background. At the end of the GIF, you see the top of Tower 12 collapsing.

Zinc-filled spelter sockets: The whys

The thing that’s confusing about the Arecibo collapse is that it shouldn’t have happened. After these incidents, as you can imagine there was a lot of finger pointing — criticisms of the NSF for cutting funding, claims that not enough supports were installed after the 1997 upgrade, there wasn’t enough maintenance—but the fact of the matter is no one could really explain why it had happened. These kinds of sockets were widely used, and there had been no recorded failure of them before Arecibo.

Testing showed that the sockets failed when the tension in the cables was less than their Minimum Breaking Strength. They should have been able to handle the load. There was no defect in design or in workmanship, no issue with the materials, no environmental effects that could be pointed to that could have caused this kind of socket failure.

That’s where the National Academies report comes in. If you’re interested in reading the entire report yourself, it’s a fascinating read, but it’s also 113 pages long, so I’ll summarize it for you the best I can here.

The report focuses in on the failure of these sockets. Each one of Arecibo’s steel cables had zinc-filled spelter sockets at both ends. This is basically a steel block with a cone shaped area where the cable is inserted. The wires are spread out, and then that cone shaped area is filled with molten zinc. The zinc solidifies, forming a solid block, and as a result anchoring the cable in the socket.

This diagram of a spelter socket from Thornton Tomasetti, which conducted a forensic investigation into the collapse of Arecibo in 2022, is very helpful if you aren’t overly familiar with this kind of socket. These kinds of sockets are widely used and don’t fail like this.

This last image is key here, so remember this. The cable is pre-stretched before it’s installed, which typically means some of the zinc extrudes from the socket. This isn’t considered a big deal, it’s standard, but it’s called “cable slip.”

Zinc-filled spelter sockets are widely used in cable-supported pedestrian bridges. The American Association of State Highly and Transportation officials, for example, states that the maximum allowable slip is “one-sixth of the cable diameter when proof-loaded to 80 percent of the cable’s minimum breaking strength.” For Arecibo’s cables specifically, this would allow for maximum slip of about half an inch.

Now, look at these photos, also from the Thornton Tomasetti report.

You can see here the cable slip on the first auxiliary cable that pulled out of its socket in August 2020. This is a slip of 1.125 inches. This kind of zinc creep allows the weight load to be transferred to the outer wires of the cable, rather than the stronger inner ones (remember the outer wire snaps we saw in the video?), and it means that the cable can bear less weight. That explains the cabe failure, but not why the zinc creep was happening.

The National Academy of Sciences report posits that the collapse of Arecibo began with Hurricane Maria. Peak winds may have been as high as 118 miles per hour, but post Maria inspections showed little damage to the telescope’s structural integrity. Nothing was recorded, but if you look at the difference in the zinc socket between a 2003 image and one from 2019, you can see significant cable slip that went unreported.

It’s not clear whether Maria contributed to the cable slip, but as the report makes clear, it should have been recorded. But even if it had been, the zinc creep wasn’t really a concern at that time. The observatory would have still collapsed, even if the post-Maria repairs had been completed, because fixing this was not on their radar.

Arecibo’s responsibility in the collapse

But the whys? Why was this cable slip so pronounced? Well, that’s what this report gets into and it’s more a process of elimination than anything else. The report posits that it was Arecibo itself that contributed to its own demise.

In other words, because Arecibo was such a powerful radio telescope, that led to a unique electromagnetic environment in which it introduced a low current into its cables. In lab testing, zinc has been found to have an elevated creep rate when electric current is flowing into it. It’s important to note that the lab conditions varied significantly from those at Arecibo, so this isn’t proof or a definitive explanation — just what might be the most likely explanation of what was so different at Arecibo as compared to all the other places these sockets have been used successfully.

Arecibo will continue to be a place for science, even without the observatory. The NSF Arecibo C3 is scheduled to open in 2025, which is a STEM focused education and research center. And scientists have proposed a replacement for Arecibo, but it has yet to be funded, so we’ll see what happens there.

Enjoyed this newsletter? I’d love your support on Patreon. It helps me keep Ad Astra free of charge.