Shortly after the collapse of Champlain Towers in Surfside, Florida, the hunt for answers began. In a rare move, the National Institute of Standards and Technology (NIST) announced that it would be taking the lead in what promises to be a long process investigating the event that took, at current count, 64 lives. The involvement of the NIST—which also led the investigation in the wake of 9/11—could signal coming changes in the United States’ building codes. 

The collapse wasn’t a product of an earthquake or other natural disaster. While many ideas have been floated to explain the event, it could be a long time before we get any solid answers. Understanding it isn't made easier by the fact that the vast majority of the remains of the structure is just rubble at this point. Yet investigators have ways of determining the factors that led to this massive, tragic disaster—even though the clues have been reduced to tiny pieces. 

The NIST declined to speak on the matter at this point, so Ars reached out to researchers in the field to get a sense of how these kinds of investigations are done. 

Some have floated the idea that a flaw in the construction or original design caused the collapse of the 40-year-old structure. Others have suggested that saltwater could have eroded parts of the structure. Others still have suggested that construction on a nearby building, or differential settlement of the underlying ground, could be factors. According to Ben Schafer, a professor in Johns Hopkins' department of civil and systems engineering, it makes sense to start off with a wide slate of possibilities—there could also be others, but it’s hard to speculate at the moment. There could also be more than one cause, he said. “It's more than a possibility—it's most likely,” he told Ars. 

Where to begin?

To start things off, the team will try to get a sense of what the 136-unit condo building was like prior to the collapse, both through paperwork and the building’s remains. According to Julio Ramirez, a professor in Purdue University’s school of civil engineering—who has been involved in numerous investigations in cases where an earthquake downed a building—NIST will likely look at the original structural drawings and the as-built drawings. 

The former document represents how the structure—composed mostly of concrete reinforced with metal rebar—was intended to be built, and the latter represents how it all came out. These documents, along with the building codes from around the time the tower was built, can help point to a cause, establish what the building’s state was before the disaster, and suggest parts of the structure that might have experienced the most stress.

“Before you even start thinking about [assessing] all of these scenarios, first you need to say, 'Well, what do we really think was the starting condition?'” Schafer said. 

NIST can also compare the raw amount of concrete and metal rebar—whether this takes the form of rubble or a fixture that’s still somewhat intact—in a spot with how much material the plans indicate should have been used in that part of the structure. For example, some have raised a concern that there appears to be less steel reinforcement than would have been normal in the 1970s in various parts of the structure. 

Ramirez also noted that, at this stage, the investigators can take a look to see if there has been any corrosion as a product of saltwater penetrating the concrete and eroding the steel reinforcements inside of it—which appears to be the case. 

On site, the team could also check to see if "differential settlement" occurred, meaning different parts of the foundation were sinking at different rates. Figuring out whether differential settlement was an issue would rely on the foundation being exposed for examination, either intentionally or as part of the site clearance. Unless NIST comes across damage to the building’s foundation that suggests it's worth investigating further, this isn’t a super likely scenario, Schafer said, adding, “That would be a surprise.”

The hunt for more evidence

The team will also likely—and it appears already has—gone to the site to collect samples and observe. According to Schafer, they’ll collect concrete and steel for later testing to determine, to the best of their abilities, the state of the building on the day of the collapse. Ramirez told Ars that the team may also remove larger sections of the building for later examination. The BBC previously reported that “[d]ebris is being taken to a large warehouse to be examined.”

Schafer said NIST will also likely talk to people who saw the collapse and comb through any video footage it can find. He noted that, when trying to develop plausible scenarios, footage can be useful. He can’t say, definitively, that NIST will use computer simulations to test its theories about the collapse, but, if it does, the footage can help confirm its accuracy. 

Depending on its findings, NIST may also attempt to learn about the construction of a neighboring residential building around the time of the collapse. There's also a lot of history to potentially comb through, such as identifying any activities performed on the building’s foundation over its lifetime, as well as looking at how nearby construction might have stressed the structure. But, Schafer noted, “[t]hat's a harder one, for sure, to sleuth out.”

In all, it’s difficult to say how far NIST will go with the investigation. If its early assessment says that there were, for example, clear construction flaws and it just took 40 years for them to come to a head, the organization may not dig much deeper than that, Schafer said. 

To the laboratory

NIST has a large and well-equipped lab that can help it take a look at the samples it took from the site. Ramirez said that it’s likely that the team collected both the rubble and the scant bits of structure that remain. “It allows them to assess if there was a significant difference in the material properties between the two sections,” he told Ars. 

NIST will likely test the remains using destructive and non-destructive methods. The team may use these methods on either the materials gathered or proxy materials of the same type. Non-destructive methods include using radar or impact echo tests; when used on larger portions of rubble—say a chunk of concrete with the reinforcements still inside—these indicate where the steel rebar is within the sample and how much of it is there. 

For the destructive tests, the team could end up using a uniaxial machine—also known as a universal testing machine—which exerts force on a material in different ways to test it for tensile and compressive strength, among other properties. This process results in the core breaking—so they can tell how durable it is. The team will most likely take a core (usually 3x6 inches) out of a piece of concrete (ensuring there’s no rebar in the core) for the sake of testing, Ramirez said. 

Concrete is not particularly homogenous: there’s no uniform dispersion of all the pieces of gravel, stone, and sand in it. As such, a core might have a larger chunk of aggregate inside of it, Ramirez said. This could impact the outcome of the tests. Similarly, the strength of the samples taken from the site might be compromised because of the amount of force involved in the building’s collapse. Usually, investigators will take multiple cores and run multiple tests, he said, in order to ensure that these differences average out. 

Going forward

Depending on the rest of its work, NIST may opt to run computer simulations using its data and observations. According to Ramirez, the team might end up using structural analysis programs—similar to the ones that he uses when designing something. Alternatively, the team could run its findings through and recreate the collapse with more sophisticated tools, like finite element analysis. This is more commonly used in the aerospace field and can account for complex geometry, irregular materials, etc. 

Software like ABAQUS can perform this task, and it appears that NIST has, at least in the past, used it. From previous findings and footage of the tower collapse, the group can use these simulations to test their ideas. “This is going to take some time, obviously. Even with the power of computers and so forth, you can imagine how painstaking this task is,” Ramirez said. 

According to Schafer, basically every major building collapse in the last 50 years has, to some extent or another, led to changes in the US’ building codes. He said he’d be shocked if there were no changes in this case—perhaps to how we prevent corrosion or organize building inspections. But, right now it’s too early to speculate. 

“The field is very committed to learning from failures. The failures are rare, which means we're generally successful, but the goal is certainly to not have anything [at] this level of tragedy,” he said.