Some of the nation’s major bridges are at high risk of being struck by ships, according to a new study by Johns Hopkins University researchers.
Hopkins on Monday released preliminary findings from their analysis, nearly one year after the Dali cargo ship hit Baltimore’s Francis Scott Key Bridge in the early morning hours of March 26, 2024, causing the bridge to collapse. The collapse resulted in the tragic deaths of six construction workers who had been repairing potholes on the bridge at the time of the collision.
Early findings from Hopkins’ assessment suggest the potential for catastrophic collisions to occur every few years, though large ship strikes should be extremely rare. Bridge design standards mandate that the annual chances of a bridge collapse from ship collision should be less than 1 in 10,000. Some of the nation’s busiest bridges that encounter the highest traffic, however, will likely be hit within our lifetime, according to the study. Engineers considered traffic from all vessels larger than 150 meters, or 500 feet long.
Among the most vulnerable U.S. bridges are the Huey P. Long Bridge outside New Orleans and the San Francisco-Oakland Bay Bridge, which are likely to be hit by a ship within about 20 years, the study found.
“With this investigation we wanted to know if what happened to the Key Bridge was a rare occurrence. Was it an aberration? We found it’s really not,” said Michael Shields, a Johns Hopkins engineer specializing in risk assessment and lead investigator of the National Science Foundation–supported study. “In fact, it’s something we should expect to happen every few years.”
After the Key Bridge collapse, Shields and his team believed chances for another such incident were higher than previously thought, and that the risk to the Key Bridge given modern ship traffic was underestimated. This led them to conduct their urgent assessment of ships around the country. Within a month of the collapse Shields’ team completed the proposal for the study.
“Five weeks after the collision, we had funding from the National Science Foundation under a program called the Rapid Response Program,” Shields told Baltimore Fishbowl. “The Rapid Response Program is a program that’s designed specifically for events like this, and in the immediate aftermath of the disaster, you can apply for funding to study the effects of that disaster, the causes of that disaster, and the broader implications of it.”
The engineering team collected and mined 16 years of U.S. Coast Guard data, including logs detailing the precise location, heading, speed, and status of every ship traveling through the country’s waters on a minute-by-minute basis. They cross-referenced variables like geolocated shipping information, hundreds of millions of data points, with port data and bridge data from the National Bridge Inventory to determine which large ships passed under bridges.
The results revealed that many bridges had vulnerabilities. Several bridges, however, had a much higher likelihood than others of a major ship collision — one strong enough to cause catastrophic damage or collapse — at least once every 20 to 50 years. The study showed that many others are likely to sustain such a ship collision within 100 years.
The Key Bridge would have placed among the 10 most vulnerable bridges in the country. The team predicted it would have likely suffered a ship collision within 48 years. The bridge was 46 years old when it fell and had sustained a minor hit from a ship before that.
“To keep our bridges safe and operational, we want the chances of a collision strong enough to take down the bridge to be less than one in 10,000 in a given year, not one in a 100. One in 100 is extremely high,” Shields said. “If I look at the San Francisco Bay Bridge, we’re likely to see a major collision once every 22 years. That is huge. We want that number to be thousands of years. That’s tens of years.”
According to the study’s preliminary results, the most vulnerable bridges are:
- Huey P. Long Bridge, Louisiana: Collision expected once every 17 years.
- San Francisco–Oakland Bay Bridge: Collision expected once every 22 years.
- Crescent City Connection, New Orleans: Collision expected once every 34 years.
- Beltway 8 Bridge, Texas: Collision expected once every 35 years.
- Hale Boggs Memorial Bridge, Louisiana: Collision expected once every 37 years.
- Bayonne Bridge, N.Y./N.J.: Collision expected once every 43 years.
- Fred Hartman Bridge, Texas: Collision expected once every 47 years.
- Martin Luther King Bridge, Texas: Collision expected once every 64 years.
- Sunshine Bridge, Louisiana: Collision expected once every 71 years.
- Rainbow Bridge, Texas: Collision expected once every 71 years.
- Veterans Memorial Bridge, Louisiana: Collision expected once every 74 years.
- Chesapeake Bay Bridge, Maryland: Collision expected once every 86 years.
- Talmadge Memorial Bridge, Georgia: Collision expected once every 88 years.
- Veterans Memorial Bridge, Texas: Collision expected once every 94 years.
- Delaware Memorial Bridge, Del./N.J.: Collision expected once every 129 years.
- Dames Point Bridge, Florida: Collision expected once every 152 years.
- Horace Wilkinson Bridge, Louisiana: Collision expected once every 198 years.
- Verrazzano-Narrows Bridge, New York: Collision expected once every 362 years.
- Golden Gate Bridge, California: Collision expected once every 481 years.
- John A. Blatnik Bridge, Minnesota/Wisconsin: Collision expected once every 634 years.
Some bridges with high traffic from large ships did not make this list because their piers are located on land, away from passing ships. These include the Duluth Lift Bridge in Minnesota and the Vincent Thomas Bridge in California.
Ship traffic is a large variable in the assessment of risk factor, but there are others, too. How wide is the channel? What is the length of the bridge span? How much distance is there between the piers? Each bridge’s individual risk factor changes based on those data points.
A layperson might read in the study that for the Chesapeake Bay Bridge a collision is expected once every 86 years, learn that the bridge was opened to traffic in 1952, and conclude that a catastrophic collision is likely to occur in the next 13 years. That would be an incorrect interpretation of the study, however, Shields said.
Shields, rounding up to 100, explained, “[Y]ou can think of it as every year there’s a, say, one in 100 chance that we would expect the bridge to be struck by a ship. So, this year, we would expect there’s one in 100 chances. Next year would be a one in 100 chance. The year after that, it would be a one in 100 chance. So, it’s not that we’re accumulating risk. It’s that every year we expect that there would be a one in 100 chance, that this would occur.”
While not every large ship collision would necessarily result in a bridge collapse, Shields said it would almost certainly cause irreparable damage and likely cause at least a partial collapse.
“There’s still a lot of uncertainty in predicting the frequency of ship collisions, even with the best data we have,” he said. “But the important point is not whether it will occur every 17 years or every 75 years. It’s that it’s happening way too often.”
Ensuring safe passage of large ships under bridges requires coordination and cooperation of a number of parties, from the shipping companies to the authority that owns the bridge and the body that runs the port. Shields’ teams’ study offers the information they need to take action to put safety precautions in place to reduce the risks of a catastrophic event like the Key Bridge collapse.
Those precautions might include dolphins in the water (the structures, not the marine mammals), fenders, and maybe a change in shipping practices. Perhaps towing very large ships under bridges might become more frequent. Those are all decisions made on the local level, depending on the individual bridge’s unique factors.
Shields told Fishbowl the study officially ends in August 2025 and they expect a final report in that timeframe. While he knows the Trump administration has put NSF’s budget on the chopping block, he has not heard definitively one way or another about this study being impacted.
“We’ve heard rumors that they want to cut from, I think it’s $9 billion down to about $3 billion for the National Science Foundation,” Shields told Fishbowl. “And I would imagine that in doing so, a program like the Rapid Response Program would, if not be cut, see severe cutbacks.”
Until he hears otherwise, though, Shields and his team are moving along to complete their work of helping the nation’s bridges increase their structural integrity and reduce chances of a shipping collision.
