On Bridge Collapses (2024)
In 2007, I inadvertently wound up researching bridge collapses after the I-35W Mississippi River Bridge Collapse in my then home town of Minneapolis, Minnesota.1
On the early morning of 27 March 2024 we saw another collapse, this in my original home town of Baltimore, Maryland. The Francis Scott Key Bridge on I-695 (The Baltimore Beltway) was hit by the MV-Dali, a ship, like many I suppose, with what might politely be described as a complex governance structure (Maersk stock was down, so one imagines Wall Street believes some liability for replacing the bridge and compensating the Port of Baltimore for months of economic losses lies there). [Photos of the MV-Dali in better days]
I’ve only been on this bridge a few times, it was narrow and high and windy and not as comfortable to drive on as a lower, flatter bridge which feels more like the ground.2 More to the point though, it is generally out of the way for a trip through Baltimore to Philadelphia or New York, from the southwest suburb of Columbia I grew up in, or the DC suburbs my mom lives in, which would be better made on one of the tunnels (the Harbour Tunnel (I-895) or Fort McHenry Tunnel (I-95() or even the northern side of the Baltimore Beltway.
And while, in contrast to everyone on TV who appears shocked and surprised that it happened, I always think it can happen here (or there, or anywhere), and one of many thoughts through my head as I walk or drive or take the train or fly or ride a boat is what disasters could befall us (getting hit by a car, a bridge collapse, a tunnel cave-in, a derailment, crashing, or sinking, etc.) I am not especially neurotic and I don’t expect it would necessarily happen, it’s just among the many possibilities that could happen.
The video of the collapse is amazing (watch it if you haven’t already), and it has garnered front-page interest here in Australia (I have done 3 interviews already), much of it with the local-angle seeking “can it happen here?" vibe. To which the answer is, probably not that very specific thing. We don’t have container ships in Sydney Harbour, the Harbour Bridge is structurally different, it is an arch bridge, and the the pylons adjacent to the base of the arch (which don’t have structural importance, but instead are more psychological) have been surrounded by abutting land, not many ships travel under the Bridge itself, but yes there are risks. A ship could crash into the arch if it were going to run aground. Cruise ships have gotten very tall, and they do travel under the bridge. I doubt they would take out the bridge deck (and even if they did collide (er, allide) with the deck, the bridge itself isn’t necessarily collapsing as a consequence.
The container ships instead use the Port at Botany Bay, and they shouldn’t have to travel under any bridges there. There are bridges that cross the George’s River — Tom Uglys Bridge and the Captain Cook Bridge — that feeds Botany Bay from the west, but ships should not be using the George’s River, aside from some barge traffic for construction or maintenance.
Hobart’s Tasman Bridge did famously collapse in 1975, discussed below. I can’t speak in detail about bridges in other Australian metro areas, a cursory examination does not turn up any similar problems. Fremantle, the port serving Perth, has some low bridges just upstream of the port, but those should only serve smaller recreational and other craft. Ships and barges hitting piers is much more of an American problem, with its many inland ports and waterways, than Australian with its coastal ports and dry inland.
That said, a ship that loses control on accident should not drift too far into a pier on a bridge of an upstream river. In contrast, a ship that is taken over by bad actors (e.g. terrorists) either via piracy or hacking, can be sent into anything and used as a weapon even without explosives. (See discussion below “Bollards for Bridges”)
Allisions
Similar Bridge-Ship or Bridge-Barge Allisions:3
AP has a list of US bridge collapses caused by ships and barges:
From 1960 to 2015, there were 35 major bridge collapses worldwide due to ship or barge collision, with a total of 342 people killed, according to a 2018 report from the World Association for Waterborne Transport Infrastructure. Eighteen of those collapses happened in the United States.
I have reposted their list and description, and added a link to the appropriate Wikipedia article if available. The most similar instances are ships colliding with bridges in the US are:
POPP’S FERRY BRIDGE: March 20, 2009: A vessel pushing eight barges rammed into the Popp’s Ferry Bridge in Biloxi, Mississippi, resulting in a 150-foot section of the bridge collapsing into the bay.
INTERSTATE 40 BRIDGE: 14 DEAD: May 26, 2002: A barge hit the Interstate 40 bridge over the Arkansas River at Webbers Falls, Oklahoma, collapsing a 500-foot section of road and plunging vehicles into the water. Fourteen people died and 11 were injured.
QUEEN ISABELLA CAUSEWAY: 8 DEAD Sept. 15, 2001: A tugboat and barge struck the Queen Isabella Causeway in Port Isabel, Texas, causing a midsection of the bridge to tumble 80 feet into the bay below. Eight people died after motorists drove into the hole.
EADS BRIDGE: 50 INJURED April 14, 1998: The Anne Holly tow traveling through the St. Louis Harbor rammed into the center span of the Eads Bridge. Eight barges broke away. Three of them hit a permanently moored gambling vessel below the bridge. Fifty people suffered minor injuries.
BIG BAYOU CANOT: 47 DEAD Sept. 22, 1993: Barges being pushed by a towboat in dense fog hit and displaced the Big Bayou Canot railroad bridge near Mobile, Alabama. Minutes later, an Amtrak train with 220 people aboard reached the displaced bridge and derailed, killing 47 people and injuring 103 people.
SEEBER BRIDGE: 1 DEAD May 28, 1993: The towboat Chris, pushing the empty hopper barge DM3021, hit a support tier of the Judge William Seeber Bridge in New Orleans. Two spans and the two-column bent collapsed onto the barge. Two cars carrying three people fell with the four-lane bridge deck into a canal. One person died and two people were seriously injured.
Sunshine Skyway Bridge collapse: 35 DEAD May 9, 1980: The 609-foot freighter Summit Venture was navigating through the narrow, winding shipping channel of Florida’s Tampa Bay when a sudden, blinding squall knocked out the ship’s radar. The ship sheared off a support of the Sunshine Skyway Bridge, dropping a 1,400-foot section of concrete roadway during the morning rush hour. Seven vehicles, including a bus with 26 aboard, fell 150 feet into the water. Thirty-five people died.
Which is not to say this is a crisis, but it is certainly something to be aware of given the high cost of possible failures.
Also of note is the 1975 Tasman Bridge disaster in Australia.
The bulk carrier Lake Illawarra, travelling up the Derwent River, collided with several pylons of the Tasman Bridge and caused a large section of the bridge deck to collapse onto the ship and into the river below. Twelve people were killed, including seven crew on board Lake Illawarra, and the five occupants of four cars which fell 45 m (150 feet) after driving off the bridge. Hobart was cut off from its eastern suburbs, and the loss of the road connection had a major social impact. The ship's master was officially penalised for inattention and failure to handle his vessel in a seamanlike manner.
Traffic
WJZ-TV reported the Bridge had an AADT (daily traffic) of 31,000. This sounds right, it’s a 4-lane bypass and serves some of the southern suburbs and neighborhoods of Baltimore rather than being a long distance through route or serving the CBD. [For comparison, the I-35W Bridge had an AADT of about 140,000].
Our studies (see footnote below) looked at the traffic effects of the I-35W Bridge collapse at the macro and microscopic levels. Of the 140,000 trips that could no longer use the I-35W Bridge, about 90,000 turned up in increased volumes on other river crossings. This suggests about 50,000 disappeared, either trips were not made, or more likely, different destinations were found. Just as we have induced demand as the network expands, we have reduced demand when the network shrinks.
Longer distance shipments of hazardous materials that are not allowed in the tunnels will have to use I-695 the long way. Also local traffic that has to go between the two ends of the bridge (Dundalk and surrounds to Glen Burnie4 and surrounds) will be displaced to the tunnels at great inconvenience.
Rebuilding
The Twin Cities rebuilt the bridge in just over a year, through an effective Design-Build process. Infrastructure can be built quickly when it is prioritised. That infrastructure is more typically built slowly in the US is both a combination of process and money. No point in accelerating construction if the funds, or the workers, or the equipment, aren’t there.
Information rather than Disinformation
In the end, if you really want to follow the story, local news will be the best, I am just a reflector of facts on the internet (with some judgement, curation, and filtering applied). The Baltimore Banner is the best independent local news source. Social media will be mostly a cess-pool, aside from the fine folks on Mastodon.
Images
Bollards for Bridges
The wikipedia article on the 1975 Tasman Bridge disaster notes:
The engineering design of the Tasman Bridge provided impact absorbing fendering to the pile caps of the main navigation span capable of withstanding a glancing collision by a large ship, but all other piers were unprotected. This disaster shares some common features with the Skyway Bridge collapse in Florida in 1980, the I-40 bridge disaster in Oklahoma in 2002, and the Francis Scott Key Bridge collapse in Baltimore in 2024, all three involving collisions with ships. When river traffic "comprises large vessels, even at low speed, the consequences of pier failure can be catastrophic".[citation needed] In the field of structural engineering, the concept of ‘pier-redundant’ bridges refers to a bridge superstructure which does not collapse when a single pier is removed.[20] Two ‘pier-redundant’ bridges have been constructed in Australia – over the Murray River at Berri and at Hindmarsh Island in South Australia. The probability of ship impact is now regularly evaluated by specialist consultants when designing major bridges. One solution is to protect bridge piers through strengthening or the construction of impact-resistant barriers.[21]
Just as we have been on a kick of installing bollards in cities to protect buildings and even pedestrians from accidentally or intentionally wayward vehicles, we will need to do the same for bridges. So the key words here are ‘pier-redundant’, ‘fendering’ and ‘impact-resistant barriers’. We are going to go through a half-decade where all the bridges over navigable waterways are going to have to be re-evaluated, and defended through reinforcements. The risk of any particular accident may be rare, (35 worldwide in 55 years), the risk of a terrorist group doing this intentionally (either through piracy or electronic hacking or hijacking an autonomous ship) is inevitably going to go up in the absence of a response.
Additional Links
Australian standards for Ship Impact Loads (In 2017 version of AS5100)
6 missing construction workers likely died in Francis Scott Key bridge collapse
Bridge collapse disrupts key trade lane, prompting new supply chain woes
The Baltimore bridge collapse is a reminder of the scale of U.S. infrastructure [Just as a comment on this. Obviously the immediate cause was the ship and not the infrastructure, and that will be looked into. However, whether this could have been defended against remains an open question. A post-panamax container ship at speed creates an astounding amount of momentum, but perhaps enough bollards (impact resistant barriers) could have prevented the collapse - though also perhaps at the cost of sinking the ship, which is not necessarily a better outcome.]
How captains and pilots work (The Conversation)
Baltimore bridge collapse caused by DEI efforts, Utah Rep. Phil Lyman says [Morons gonna moron I guess. Even if DEI were bad, how exactly did it cause this.]
My group’s work resulting from our NSF, OTREC, and MnDOT funded study of the 2007 I-35W Mississippi River Bridge Collapse, for those interested, includes:
Carrion, Carlos and David Levinson (2019) Route Choice Dynamics after a Link Restoration. Transportmetrica B: Transport Dynamics. 7(1), 1155-1174. [doi]
Danczyk, Adam, Xuan Di, Henry Liu, and David Levinson (2017) Unexpected versus Expected Network Disruption: Effects on Travel Behavior. Transport Policy. 57, 68–78. [doi]
Di, Xuan, Henry Liu, Shanjiang Zhu, and David Levinson (2017) Indifference Bands for Route Switching. Transportation. 44, 1169–1194. [doi] [full-text view only]
Zhu, Shanjiang, David Levinson, and Henry Liu (2017) Measuring Winners and Losers from the new I-35W Mississippi River Bridge. Transportation. 44, 905–918. [doi] [full-text view only]
Carrion, Carlos and David Levinson (2011) A Model of Bridge Choice Across the Mississippi River in Minneapolis. (115-130) in Network Reliability in Practice (ed. by David Levinson, Henry Liu, and Michael Bell) Springer.
Xie, Feng and David Levinson (2011) Evaluating the Effects of I-35W Bridge Collapse on Road-Users in the Twin Cities Metropolitan Region. Transportation Planning and Technology 34(7) 691-703. [doi]
Zhu, Shanjiang and David Levinson (2011) Disruptions to Transportation Networks: A Review. (5-20) in Network Reliability in Practice (ed. David Levinson, Henry Liu, and Michael Bell) Springer.
Zhu, Shanjiang, Nebiyou Tilahun, David Levinson, and Xiaozheng He (2011) Travel Impacts and Adjustment Strategies of the Collapse and the Reopening of the I-35W Bridge. (21-36) in Network Reliability in Practice (ed. David Levinson, Henry Liu, and Michael Bell) Springer.
Zhu, Shanjiang, David Levinson, Henry Liu, and Kathleen Harder (2010) The Traffic and Behavioral Effects of the I-35W Mississippi River Bridge collapse. Transportation Research part A 44(10) 771-784. [doi]
I also had a series of summary blog posts in Streets.MN on the net effects, which comprise a chapter of the 2nd Edition of The Transportation Experience.
Not that I have an irrational fear of heights, I fly without problem and am happy to use an elevator to go to the upper floors of a tall building and look out their secured windows. I think it is quite rational to avoid unnecessary risks like standing adjacent to the edge of a cliff protected by a less than waist high barrier.
“It should be recognized that the USCG refers to ship and barge accidents as “allisions”, which is technically correct as it involves an accident between a single moving object (ship/barge) and a stationary object (bridge). Whereas a “collision” is technically an accident between two moving objects.”
[Of course if we are going there, the earth has been in motion at least since Galileo. And I had to study dynamics in class, so stationarity is relative.]
We always made fun of “Glen Burnie” as kids, because the name sounded funny.
I was under the impression that large ships were escorted by tugs (whose pilots knew every inch of the harbor) when entering or leaving inland harbors. Why wasn't the Dali escorted by tugs?