NTSB recommends risk assessment for 68 bridges for vulnerability of collapse

[Big John]

https://www.cnn.com/2025/03/20/us/baltimore-bridge-collapse-ntsb-findings/index.html

Sixty-eight bridges across the US should be assessed to see if they are at risk of collapse if hit by a ship, transportation safety officials found, while urging the bridges' owners to undertake immediate vulnerability assessments. The four urgent safety recommendations are part of the National Transportation Safety Board's ongoing investigation into the collapse of the Francis Scott Key Bridge in Baltimore last year.

In the early hours of March 26, 2024, the container ship Dali lost power after leaving the Port of Baltimore and struck a pillar of the Key Bridge, causing it to collapse, killing six construction workers who fell into the Patapsco River. The Dali suffered a pair of catastrophic electrical failures minutes before the crash, according to a preliminary report released by the NTSB last May.

The Key Bridge was above the acceptable level of risk based on guidance established by the American Association of State Highway and Transportation Officials, or AASHTO, NTSB officials said. But the owner of the bridge never evaluated that risk.

"The Maryland Transportation Authority never ran the calculation on the Key Bridge," NTSB Chair Jennifer Homendy told reporters Thursday. "Had they ran the calculation on the Francis Scott Key Bridge, the MDTA would have been aware that the bridge was almost 30 times greater than the risk threshold AASHTO sets for critical, essential bridges."

The NTSB identified 68 other bridges in 19 states spanning waterways frequented by cargo ships that, like the Key Bridge, were built before 1991 and do not have a current vulnerability assessment. Owners of the bridges that have higher than acceptable risk ratings should create a plan to reduce that risk, Thursday's findings say.

The Golden Gate Bridge in California; Brooklyn, Manhattan, Williamsburg, George Washington and Verrazzano-Narrows bridges in New York City; the Walt Whitman and Benjamin Franklin bridges in Pennsylvania; the Sunshine Skyway in Florida and the Mackinac Bridge in Michigan all made the list.

"A risk level above the acceptable threshold doesn't mean a collapse from a vessel collision is an absolute certainty," Homendy said. "What we are telling bridge owners is that they need to know the risk and determine what actions they need to take to ensure safety."

[SectorZ]

I wonder if we had a similar discussion after the Sunshine Skyway Bridge collapse in 1980.

[Max Rockatansky]

This strikes me as an overreaction.  It isn't as though Key Bridge-like events are a common occurrence.  Some of these bridges do have mitigative measures currently in place for impact events.

[wanderer2575]

And is the NTSB making any effort to convince government to make available the funding needed to implement these risk reduction plans?  Or is it once again just a think tank churning out opinions?

[BrianP]

I wonder if we had a similar discussion after the Sunshine Skyway Bridge collapse in 1980.

I think there's a pertinent sentence in the WTOP article. That's not in the CNN article. The CNN article mentions bridges before 1991.  But leaves out why that year was pertinent.

That threshold is set by the American Association of State Highway and Transportation Officials, according to the NTSB, which first published the vulnerability assessment calculation for new bridges in 1991.

So they may have had discussions back then.  But they didn't have the calculation to quantify the risk until a decade later. That calculation was possibly a byproduct of that earlier incident.

https://wtop.com/dc-transit/2025/03/watch-ntsb-to-provide-update-on-baltimore-key-bridge-collapse-investigation-nearly-1-year-later/

Another good quote from the WTOP article:

"What's frustrating is, not only did MDTA fail to conduct the vulnerability assessment on the Key Bridge, they did not provide, nor were they able to provide the NTSB with the data needed to conduct the assessment."

The data the Maryland Transit Authority could not provide included the speed of vessels passing under the bridge, vessel loading characteristics, water depths, environmental conditions, bridge geometry and capacity of the bridge piers.

"We asked for that data — they didn't have it. We had to develop that data ourselves," Homendy said.

MDTA fail indeed.

[jeffandnicole]

I wonder if we had a similar discussion after the Sunshine Skyway Bridge collapse in 1980.

Being the current Vulnerability Assessment guidelines were developed in 1991, it's a clear Yes. 

The issue is those guidelines only impacted bridges engineered and designed from that point forward.  Bridges built prior to that time, or even bridges under construction at that time, were not required to undergo the assessment.

Of the bridges on the lists in my area, the Delaware Memorial Bridge are 2 of 4 bridges that have undergone assessments and have active or near active plans to migrate concerns. In this case, they are actively building dolphins near the piers; the construction equipment can be seen as you drive over the bridges.

Of the 68 listed in the report, the 4 DRPA bridges (3 listed under Pennsylvania, 1 listed under New Jersey for some reason) actually have pretty good protection, either with islands or land around the piers or land piers jutting out close to the bridge piers. Only the Ben Franklin Bridge appears to have the NJ-side pier exposed to down-stream ships.

Interesting to note that the 2 Burlington County Bridge Commission bridges are *not* listed on the report, even though they're in navigational portion of the Delaware River, between the DRPA Bridges and the NJ/PA Turnpike Delaware River Bridge which were listed.  I wonder if there was an oversight within the recommendations as they are lift bridges that, in their closed positions, wouldn't have the 80 feet of clearance specified as needing additional reviews.  There's also a NJ Transit rail bridge that was similarly omitted also.

[kphoger]

https://www.cnn.com/2025/03/20/us/baltimore-bridge-collapse-ntsb-findings/index.html

"Had they ran the calculation on the Francis Scott Key Bridge, the MDTA would have been aware that the bridge was almost 30 times greater than the risk threshold AASHTO sets for critical, essential bridges."

So they may have had discussions back then.  But they didn't have the calculation to quantify the risk until a decade later.

I have to ask:  is "30 times greater" even a meaningful comparison?  As in, we're not talking about the number of peanuts in a bag.

[PColumbus73]

This strikes me as an overreaction.  It isn't as though Key Bridge-like events are a common occurrence.  Some of these bridges do have mitigative measures currently in place for impact events.

Especially where the Sunshine Skyway and George Washington Bridge was mentioned.

[kalvado]

https://www.cnn.com/2025/03/20/us/baltimore-bridge-collapse-ntsb-findings/index.html

"Had they ran the calculation on the Francis Scott Key Bridge, the MDTA would have been aware that the bridge was almost 30 times greater than the risk threshold AASHTO sets for critical, essential bridges."

So they may have had discussions back then.  But they didn't have the calculation to quantify the risk until a decade later.

I have to ask:  is "30 times greater" even a meaningful comparison?  As in, we're not talking about the number of peanuts in a bag.

You have to have some numerical estimates to justify the project and costs.  Same with seismic events, for example.
It makes more sense to spend money to protect a bridge with active sea traffic than the one which sees 2 small boats a year. And you spend much more on seismic preparedness in California than what would be justified in NY or MA... Numbers would look something like "1 estimated collision in 10 years, 1 in 100 may damage the bridge" - and suddenly those numbers are meaningful... 

[kphoger]

Numbers would look something like "1 estimated collision in 10 years, 1 in 100 may damage the bridge" - and suddenly those numbers are meaningful... 

I just have zero context for judging whether "30 times greater" is even a meaningful difference, whether it's a big difference or a small one.

Take crime statistics as an illustration.  In the state of New York in 2022, Albany had 10 times the robbery rate (175.4 per 100k) as did Albion (33.5 per 100k).  But I'd argue that's a meaningless comparison to make.  But because Albany is much more similar in population size to Wichita Falls (TX), then I'd say it's more meaningful to say Albany had 3.5 times the robbery rate as Wichita Falls.

If the threshold set by AASHTO is incredibly low, then "30 times greater" might not actually be all that much higher.  Or, for all I know, there are scores of other bridges out there that are also "30 times greater" than that threshold.  That's why I was asking if it's even a meaningful comparison to make.

[hotdogPi]

Albany had 10 times the robbery rate (175.4 per 100k) as did Albion (33.5 per 100k).

This is only 5 times.

[kphoger]

This is only 5 times.

I also got my starting numbers wrong. Let's try this again:

236.5 = Albany, NY
35.6 = Albion, NY
49.4 = Wichita Falls, TX

There we go.  My point stands.

[kalvado]

Albany had 10 times the robbery rate (175.4 per 100k) as did Albion (33.5 per 100k).

This is only 5 times.

I think albany is improving, they should be reaching 10x now.

If the threshold set by AASHTO is incredibly low, then "30 times greater" might not actually be all that much higher.  Or, for all I know, there are scores of other bridges out there that are also "30 times greater" than that threshold.  That's why I was asking if it's even a meaningful comparison to make.

There has to be a threshold of some sort. You need to define where to spend money, and how much to spend. Once resources are limited, you need to prioritize. This is what actuary math is about.
As a semi-wild guess: if all 68 bridges in this list score the same as the collapsed one, and that is once in a 30 year thing among 68 at risk (wasn't there similar event in 1990s?)- individual bridge has 1 in 2000 years risk of failure.  If that is 30x too high, maybe the goal is 1 in 100k years for each individual one?
And if the cost of collapse. with casualties, cleanup and rebuild, is $10B...  with 1 in 2000 risk for next estimated 20 years- that is 1% total risk of loosing $10B, so spending $100M makes sense - but $1B remediation project isn't worth it.  That is, of course, a simplified one...

[jeffandnicole]

And then you get the media to post very mis-leading headlines, such as this from nj.com: "N.J. bridges that should be checked out for risk of collapse."

https://www.facebook.com/photo/?fbid=1065282918975493&set=a.373809851456140

[Scott5114]

And is the NTSB making any effort to convince government to make available the funding needed to implement these risk reduction plans?  Or is it once again just a think tank churning out opinions?

Uh, the NTSB is the government. They are limited in their ability to influence funding, though, since that's determined by the legislative branch and NTSB is part of the executive branch, so they are supposed to be independent of each other.

[Moose]

The Big Mac is on the list.. but.. I would not worry about it.

The lakes are a whole different deal.. plus it's not on a port exit. Requiring many many hours from most ports to pass, plenty of time for a problem to surface before hand.

Plus the lakes have plenty worse narrow spots then the passing under the Big Mac.