The Grand Unified Theory of Rogue Waves | Hacker News Reader

The Grand Unified Theory of Rogue Waves

quantamagazine.org

136 pointstheafh41 comments

If anyone wants to see a clear video of a rogue wave in the wild: https://www.youtube.com/watch?v=uK_4V3zqAvg

I wish people walking around with personal video cameras in their pockets were a thing at the turn of the millennium. I was on the bridge of an aircraft carrier going around Cape Horn, taking green water over the bow. Not once, but many, many times.

> [a better model] paving the way for machinery that could, for instance, scan the ocean and notify ship captains that they face a 13% chance of running into a 30-meter wave in the next 15 minutes.

What sorts of actions would captains take under these circumstances? Rogue waves are so large and powerful--would any action be enough to prevent losing a ship?

You'd want to orient the ship to hit it bow first probably, rather than on the side where it could roll or punch a hole in the ship. But what Captain in a serious storm is not going bow first through the waves anyway? I don't know much about the industry, so that's an honest question - maybe deadlines and route optimization still plays a role and they will often sail at an angle to the waves, I doubt they would ever sail near 90 degrees in any case, but maybe you could change a 30 degree angle to closer to 0 if you had warning.

It's absolutely normal to steam with little regard to wave direction in mild seas. Merchant ships typically follow great circle routes with as little deviation as possible.

Right, but in a serious storm?

I think the point of a rogue wave is that it can form even when there's not a storm. They seemingly come out of nowhere.

Not really, they're still proportional to the average wave height. So while you could always be at risk for rogue waves, the sea has to be quite rough before a large vessel need be worried about them.

I didn't see anything in the article that implied the rogue waves were proportional to the average. In fact the thing that makes them rogue is that they are many standard deviations away from the average.

This is not the first article I've read on the subject, it's my understanding that rogue waves are built from regular waves, therefore they're proportional - just many standard deviations away as you say. You can get a 20 meter wave in the midst of 10 meter waves, but you're not going to find that in 2 meter waves, rounded to an appropriate number of decimal places.

The whole point of the article is that rogue waves aren't just built from regular waves syncing together...

They don't come out of math...

Don't they?

https://youtu.be/iWKFPTgkpXo

That's a 90 foot wave in a wave tank basically coming from math

Do you have a citation for that? The definition of a rogue wave is a function of existing conditions, but that doesn't exclude a rogue wave in calm conditions; it just requires a greater height in rougher seas.

Intuitively I would expect rogue waves to be more likely in rougher seas, but it's another thing entirely if the physics of rogue waves requires rough seas locally. I don't disbelieve you, but an explanation of that claim would answer a ton of questions about how rogue waves form (answers missing from the Wikipedia page, which seems more inconclusive) and a page worthy of bookmarking.

EDIT: Your claim seems consonant with the "linear" theory of rogue waves, but as the Quanta article explains that theory is contested by proponents of the "non-linear" theory.

I think I see the confusion, rogue waves, by definition, are just waves much larger (twice as large, again from the definition) than the surrounding ones, sufficiently so that they're impossible under the old wave models. So you can obviously get them in any conditions. But you're not likely to get one that can sink a large container ship in otherwise calm seas (although it's not in theory impossible), which is what I was talking about if you go back to my original comment. The height is still proportional to the other waves.

If the wave is big enough, bow-first won't save you.

If you go over the wave, half your ship ends up cantilevered over the peak. A ship isn't built to hold half its weight in the air like that. It will break.

If you go through the wave, the upper decks get hit by a wall of water. The windows blow out, the bridge is destroyed, the computers are toast and the ship is swamped.

Force shields it is then.

Agreed, that if the wave is big enough you're toast anyway, but bow-first still gives your best chance as I understand it. So if you had warning you'd want to get your crew into less vulnerable positions where they can evac more easily, and ensure you're facing bow-first into the waves.

All that is left is to go under the wave.

FWIW, this is exactly what surfers do ("duck diving").

Why don't we design submersible container ships? Are the energy costs of displacement while moving entirely underwater too expensive?

Perhaps the density of the cargo is, on average, too low. The amount of ballast needed would make it impractical.

It would definitely use more energy, and container shipping is all about efficiency. It's a non-starter.

are you sure? wave drag is a big part of the total drag ob a ship. which is also why dolphins go pretty fast with much less than one horse power.

Many military subs are actually faster submerged than surfaced.

The wave drag (or wave resistance, to disambiguate from the shock wave drag of supersonic aerodynamics) referenced in sibling comment is the energy contained in the waves created by a surface vessel. The details that you'd have to understand to optimize a hill shape are quite complicated, but for a simple surface vs sub comparison it's enough to know that surface waves exist and that they contain energy that is projected away. A sufficiently submerged vessel does not create those.

Submarines require a fantastically higher amount of maintenance to keep running.

After reading Outlaw Sea I think even the maintenance that is required on container ships is frequently skipped or corners cut, at least by smaller operations.

Cargo handling time at port is a big cost factor, it would be very hard for a sub to provide comparable access for loading/unloading.

Are you suggesting a submersible that might run underwater only temporarily, during the storms, to avoid the undulating surface?

Hitting the swell bow first will not necessarily help you. See https://www.youtube.com/watch?v=A2KqofR05TE for a video of a boat actually hit by a rogue wave. They are riding the swell correctly - but the wave hits from the side.

Note that a swell and a current are known conditions that make rogue waves more likely. And it is not uncommon for rogue waves to be moving in a different direction than the swell.

Someone else also commented that it won't save you if the wave is big enough. I'm not contesting that. I'm just saying that gives you your best chance.

You ignored the point that rogue waves often move in a different direction than the swell. So orienting by the swell doesn't help you with a rogue wave.

I'm not sure if I just missed it because I was busy, or if you edited the comment afterwards, but that's a good point. If the rogue wave comes from a different direction then orienting with the rest of the waves won't save you. In that case maybe if you could get some warning in time it could help. You might not be able to get to 90 degrees, but maybe you won't get caught flat out on the side either.

Still even that may not be enough to save you. The ship that went down in the North Atlantic was hit front on and sunk.

In the sort of seas that produce rogue waves, the direction of the rogue wave is also less predictable.

Batten down the hatches.

Or maybe a herd strength thing where one ship detects and can send out an alert for others to redirect around the path.

You could make sure all your people are awake and ready to get in the lifeboats if the ship breaks up. Definitely make sure no one is way down in the hold where they'll drown before they get on deck.

this seems like an insurance problem, rather than a technical one. rogue waves appear to be rare enough that spending money on prediction or detection might outweigh the cost of simply insuring against catastrophic loss

Tell that to the people on the ship that it happens to!

Granted, it's usually survivable. Decades ago a friend of mine on a containership experienced a rogue wave with no more harm than being terrified by "either the last thing I was ever going to see or the most majestic thing I'd ever see" but it doesn't always end that way.

It is both.

Ships used to be built to the largest waves that were conceivable. We underestimated the forces in a rogue wave by a factor of more than 5. Ships today are built stronger, we are careful about seas where rogue waves are more common, and insurance now knows to take rogue waves into account in their policies.

Better predictions as well would be good.

Does this have anything to do with Markov Chains? Because it's smelling like that type of problem.

The verbiage in the article is a little too "wanky" for me to tell.

If you had to reach into your bag of probability concepts to pick the most closely related one, it would be "Large Deviations", which comes up for engineers in the context of information theory.

Basically, as explained in the article, many kinds of rare events (like, averages attaining values far from the population mean) can be characterized without knowing a lot of problem-specific details.

For instance, if the typical value for an average of 1000 positive numbers is 10, and you want to know the chance of an outcome greater than 100 (a "large deviation"), you can pretty much just calculate the chance of the outcomes very near 100, without caring about 120 or greater. That is, it turns out that the larger outliers carry very little probability, although that's not obvious.

This phenomenon comes up a lot and goes by several names. It was known to Laplace and there's an approximation used commonly in that area bearing his name. It's also related to the notion in information theory of the "typical set".

Similarly, you can, in some ways, characterize the realizations that cause these extreme values. (As opposed to just computing their probabilities.)

For instance, if I remember right, the typical realization that gives rise to a large average has just a few larger-than-expected values and a vast majority of about-average values, rather than every value being slightly too large. This is a highly general property.

Reply on news.ycombinator.com