After reading this, I think not. But I’m not an engineer. (Gift link) Created a new thread because the other was breaking news.
How could something traveling slower than a casual bike rider cause such a devastating impact? The answer lies in its mass: roughly a third to a half of the Empire State Building.
My bold.
Interesting article. Much to appeal to math nerds.
Our local (San Francisco) news station did a little info bit about the bridges (Golden Gate, Bay Bridge) that are the best known. I believe they said that the Golden Gate Bridge has the most stringent protection against ship crashes of any bridge in the US (or something like that). There is a very large steel bunker filled with sand around the south pier (the north pier is actually on land) that should protect against any similar kind of accident. Now, I know these reporters are not great with details, but I assume someone more knowledgeable than they wrote that stuff out for them to say.
Could any bridge have withstood Baltimore collision?
Well the Sydney Harbour Bridge could, but that’s because it spans the shorelines and doesn’t have any pylons within the shipping lane.
Well, it’s about as good as you could expect from a general news publication but the focus on F = m*a misses the point that the key relationship is momentum and resilience (or elasticity). There is a static force that will certainly cause a failure, but even short of that force if the ship can continue impacting the support to the point that it starts to deflect such that it becomes unstable or starts to buckle, it will still fail.
Protection for the bridge supports could, in theory, be made robust enough to have resisted this impact, and it is possible that the extremely robust supports on the Golden Gate Bridge or others could have survived. But the Francis Scott Key Bridge was built in the ‘Seventies when container ships were a tenth the size and capacity than the large vessels like the Dali are now, and may (or may not) have been sufficient at the time but never revisited with respect to modern container ships.
I’ll note that the Golden Gate Bridge enters almost directly into the Pacific Ocean, and San Francisco Bay is one of the most challenging tidal current environments in North America. (Having kayaked under the bridge and all the way out to in-line with Point Bonita Lighthouse, as well as having sailed in an out of the harbor in sailboats of various size, I can personally attest to this) so it makes sense that they would build highly protected supports. I am not at all familiar with Baltimore but where the Key Bridge is located is deep within Chesapeake Bay so I wouldn’t expect big ocean wave interactions or the kind of radical shifts and turbulence that are frequently experienced in San Francisco Bay.
Stranger
Yeah, this is the most salient fact.
I thought it was an excellent article, even if the physics were slightly simplified. It helped me put things in perspective, except perhaps for the confusing reference to the gravitational force on 100 female whales, which I find hard to picture for one reason or another.
Reading the article comments, I would agree the problems are not just bridge design and even mitigation with dolphins and other such safety mechanisms. Mandatory use of tugboats in case of propulsion failure would seem to be very important, and of course limiting ship sizes might help prevent future tragedies. So might the insurance battles over responsibility and the forthcoming details of what actually happened.
I think people have a lot of mistaken ideas about the capabilities of harbor tugboats used for handling container vessels. Tugs are used to essentially push or pull the a container vessel or other large vessel into or away from quayside, and the tugs that are typically used for this are smaller, designed for high maneuverability rather than maximum bollard pull. They push the vessel very slowly in position as directed by the harbor pilot, and docking is often the work of hours of coordinated activity while the vessel is essentially not underway (moving at a fraction of a knot). Even if there were a larger “pusher” barge tug available (and the hull of the container vessel could withstand the impact) it probably couldn’t have gotten in position in time to redirect the Dali because there was just too much forward momentum heading toward the bridge pier. Nor is anyone going to accept “limiting ship sizes” because the viability of global shipping is highly dependent upon maximizing cargo and minimizing costs.
I don’t know that I agree with the argument that sufficient reinforcing and protective accommodations couldn’t have protected the bridge. With enough reinforced structure any size ship could be stopped, although at significant cost. This guy claims to be an experienced civil engineer who has worked on bridge pier protection and makes the argument that with sufficient reinforcement the bridge could have resisted the impact:
Stranger
A container ship isn’t an irresistible force, so all a bridge requires, to be able to withstand such a collision is for the piers to be sufficiently immovable objects. That’s not impossible, but it might be difficult and/or expensive - for example placing the piers on substantial artificial islands would prevent it - that is, if they were substantial enough that the ship basically runs aground before it hits the structure.
There might be reasons why that’s incredibly difficult, or nearly impossible - such as that any large enough artificial island would present its own hazard to shipping, or would not fit in the channel without restricting the navigable part. But the answer to the question ‘can anything stop a container ship that is in motion?’ is definitely not ‘no’.
One thing that made this collision particularly catastrophic was the continuous truss structure - if part of that comes down, it’s all coming down; there are ways to construct bridges with sections that are more modular, comprising a series of individual trusses, so a collision might take down any part that is directly supported by the pier that is destroyed, without pulling the other sections off their piers. There might be practical reasons why this bridge in this location couldn’t be built that way though.
I know nothing about tugboats. You’re probably right. But a lot of responses to the article are from people who claim expertise such as navy officers or port people. They may or may not have this expertise, but they likely know more than I do about it. They claim large navy ships require tugboats, but no doubt there are many differences in commercial ports, with much bigger ships, things like bridges, and considering local factors and changes over time.
The Navy also doesn’t care what it costs to move their ships in and out. The shipping industry overall is super super cost conscious / cheap.
If all cargo ships had two engines instead of just one, dual redundant rudder systems, etc., a LOT of the total shipping accidents would be preventable.
I think one item of note is the widths involved here. This isn’t a case of a panamax ship barely squeezing through the Panama Canal. The Dali is 158 feet wide. The distance between the two supports on the Key bridge was 1200 feet. That’s a path around 7.6 times the width of the container ship. That’s the equivalent trying to maneuver a tractor trailer between two barriers five freeway lanes apart. At nine miles per hour. The harbor authorities or whatever you call them had to think it would take a spectacular level of incompetence to not be able to thread that needle.
Incompetence or failure.
The problem with a hefty percentage of ship collisions / allisions is not human error in steering / piloting.
It’s malfunctions that will take 10s of minutes or hours to repair while only 2 or 5 minutes away from a crash, and insufficient or malfunctioning relevant backup system(s) installed. Or worse yet, no relevant backup system(s) installed at all.
If you go to Investigation Report (ntsb.gov) and click on [Marine] you’ll see a map and a listing of recent completed investigations. From there you can also dig into NTSB’s extensive history of marine accidents.
The vast majority are mechanical failure that leads to a crash before there’s time to recover. Yes, plus some incompetent responses to mechanical failure, plus some plain old careless or incompetent crewing and Captaining. But if major failures of essential but non-redundant systems didn’t happen, neither would most accidents.
As long as international law and flags of convenience dominate the commercial shipping industry, the vehicles will remain highly accident prone. By design. Design to cost that is.
Of US Navy ships only supercarriers approach the weight of even middling supermassive container ships like MV Dali, and as you note, the cost is basically a part of their operations funded from a nearly bottomless budget, not drawn against narrow operating budgets.
Unless they had redundant boilers, a loss of engine power would still result in loss of steerage. And for an industry as cost-conscious as intercontinental shipping, the costs of building those into new ships and the associated maintenance is a nonstarter unless you are willing to pay the premium on cheap goods you buy at Wal-Mart or from Amazon.com.
Stranger
That just means that the intervention should have been sooner. The ship should be correctly steered to clear the bridge, far enough out from the bridge that if there’s any problem, there’s time for tugboats or other accommodations.
There was four minutes between loss of the engine and the impact with the bridge pier. That may seem like a lot of time to people unfamiliar with large vessels but again these are giant hulks of mass that are about as easily redirected as a herd of elephants. And there is no unlimited quantity of tug boats (and skilled crew) to escort large vessels out to sea. It is really easy to sit behind your keyboard and say what “should have been done”; a lot more difficult to formulate and implement practicable solutions. Frankly, the focus here should be on why the engine failed without notice, whether that is due to poor maintenance, contaminated fuel, or some other issue, not on febrile notions of nonexistent fleets of tugs providing escort to every large vessel entering or exiting waterways or the expectation that someone will come up with hundreds of millions of dollars to reinforce bridge supports to resist the (thankfully rare) impact by modern giant container vessels.
Stranger
'Zactly.
Ultimately from a systems safety perspective, when we have large non-redundant machines operating near critical infrastructure that cannot withstand an impact we are making a societal decision that cheap shipping is more important than periodic statistically inevitable bridge destructions.
The three possible fixes are:
All ships are tug escorted (actually better yet tug powered and tug maneuvered) from the seaward side of anything they could destroy by impact until tied up. OR
We rebuild all bridges and waterfront whatnot to be collision impervious for ships significantly bigger than present size and commit to improve everything before allowing even bigger ships than whatever the current design standard is OR
We rebuild the worldwide shipping fleet to be sufficiently redundant that these sorts of failure are 1E-9 likely per transit, not 1E-4 likely as now.
All of those are very expensive, and the last one requires worldwide standardization while the other two only require local or country level action.
Still, ain’t gonna happen on any timescale measured in units much smaller than “lifetimes”.
Wikipedia says the draft of the MV Dali is 49 feet. The picture above says the depth at the pylon was 20 feet. Even assuming that the Dali was using well under its max draft, it was fully loaded and I’d think it would be well over 20 feet. So how did it get that close to the pylon in the first place?
I watched the videos you linked above. What I am unclear on is what a tug can do to a ship underway. He mentions the port at New Haven which actually does use tugs to escort ships in and out. So, seems possible but I truly do not know. He mentions stopping the ship is basically impossible for tugboats at that point but it seems the ship only need to be nudged back to port a little…just enough to clear the bridge base. Can a tugboat do that to a ship that size running that speed? I have NO idea.
As for distance traveled, which is mentioned in the videos, ISTM the distance to the FSK bridge is really pretty short and doable from that port. All the way to the second bridge and the sea would be silly and take hours. Better to station two tugs at the second bridge. No need to follow the ship the whole way. I agree calling for tug boats with five minutes to hit the bridge would never work…no chance in hell they’d get there in time unless there were already right there.
As for cost, let’s say this adds $20,000,000 to port operations. If you have one cargo ship per hour moving through that would add $2,300/ship to the port cost. I think for ships like these that is change they find in their couch. You can fiddle with the numbers as you like but the extra cost does not seem unbearable even to cost conscious shippers.
And hey…more jobs for the port.
All that said, I think it would probably make more sense to build defenses for the bases of bridges. Probably expensive upfront cost but once done they just do their thing with little ongoing cost.
I 100% admit I have no experience in this stuff. Just thinking out loud and curious.
According to their website the Port of Baltimore, Maryland
“serves over 50 ocean carriers making nearly 1,800 annual visits.”
Which is just under 5 per day.
So the service cost is not going to change, because the tugs will always be on station, but the number of movements has fallen by 80%.
Having been in the import/export business for most of my working life, stevedoring companies know how to turn a service cost into a profit.
I don’t see how using tugs solves the problem. It seems to me it just shifts the danger from “cargo ship’s steering system fails at an inopportune time” to “tug’s steering system fails at an inopportune time”. Is there a reason to think that tugs are more reliable than larger ships?