Big data at sea: How the Royal Navy charts the world's oceans You may think of a warship as a vessel that sails the Seven Seas, bristling with missiles and guns, ready to deal out death and destruction to Her Majesty's enemies. In fact, warships do many other jobs too – such as HMS Enterprise's routine but vital task of seabed surveying. Your correspondent spent most of last week aboard …
So the argument goes:"I'll plough this field with my spade, because all 10 of my tractors are stealth tractors that I don't want the next farmer to know about."
No. To use your analogy, 'I'll use my tractor to do the ploughing because my fleet of race cars would be bloody useless for the job'.
Replace "tractor" with "purpose built survey ship", and "race cars" with "state of the art, nuclear powered, submarine" if the metaphor is a bit murky.
Semtex451,
They ain't stealth tractors. The Astute class cost something like £2 billion each. If we'd not buggered about delaying the purchase and then thinking about changing the design, and so delaying them again, then the carriers wouldn't have cost all that much more than the Astutes. Despite being somewhat larger...
Nuclear submarines are horrifically complex, and have to be built to much higher tolerances, which also makes them much more expensive to maintain. They're also busy doing other things. Given the state of our relations with Russia I would be surprised if one of them isn't wandering round in places where it shouldn't be, poking it's nose where it's not wanted - as they did in the Cold War. And I think with subs you'd expect them to spend as much time under maintenance as they do at sea. Remember we have 4 Trident boats just to make it possible to keep one at sea at all times. According to the Navy's calculations an absolute guarantee of one nuclear sub permanently at sea requires 5 boats and with only 3 it's impossible.
Oh and also the stealt thing is really important. During the Cold War the RN put in massive effort to stop the Soviets from ever tracking a Trident boat. At least they said publicly they believed they'd managed it. That's because once you've tracked a boat you can record and build a profile of the noise it ouputs, which then means it's subsequently easier to track.
@ I ain't SpartacusVery many thanks for fully and properly answering my innocent question.
No problem. I forgot to mention a couple of other things. One's been covered by someone else - if the submarine isn't near the surface it can't use GPS. If it is, then it has no advantage over a surface ship.
The other is that the ship can regularly drop probes to measure water conditions at various levels. As well as having that device to do it that can't be deployed in rough weather. And that info on temperature, salinity etc is vital to an accurate survey. Whereas a sub is only at one depth at any one time. Although obviously it can change that, but that would affect the survey data.
Finally there's more room on a survey ship. For general gear and comfort of the crew.
Having said that, as I understand it, at least one of our subs did participate in the search for MH370. But I don't know if that was helping with the survey work, or searching for the sonar pinging of the black boxes in the first few days. I suspect the latter, because it was in the early days. We also sent one of the survey ships for some of the search operation.
"I'm very glad I asked, it was was worth the downvotes and snarks to get such illuminating answers as yours and rg287's. To my small mind it isn't such a stupid question."
Well, I hope I stopped at snark. Still it just goes to show, if you want to learn something on the internet, you'll get more answers if you post something wrong, rather than asking a question.
"And I think with subs you'd expect them to spend as much time under maintenance as they do at sea. Remember we have 4 Trident boats just to make it possible to keep one at sea at all times. According to the Navy's calculations an absolute guarantee of one nuclear sub permanently at sea requires 5 boats and with only 3 it's impossible."
That's odd. I thought the whole point of a nuclear sub was that it could stay submerged for weeks / months on end (basically as long as food/drink supplies last). Does it REALLY take a year to do maintenance on a sub that has been out 'in the field' for a year? And even if the operation:maintenance ratio is close to 1:1 then surely with 2 subs you're almost covered and 3 should be fully covered? If they need 5 subs to absolutely guarantee that 1 boat is operational the operation:maintenance ratio must be far below 1:1
jmch,
I don't know actual details. And obviously the nuclear deterent is an extreme example of numbers of ships required to fulfill a role. But In the 30-40 year life of a boat it's going to have several year-long (or even multi-year) periods out of commission for major systems upgrades. They used to have to refuel the reactors, or just replace the whole damned thing, but I think Astute is supposed to have its entire lifetime's fuel aboard - so that's one less long refit.
You've also got shorter periods of maintenance, probably because they can run for so long without it. Systems will fail the longer your cruise is, due to constant use, and you can do some running-repairs or switch to back-ups. Plus there's probably all sorts of upgrades to onboard systems that are quick, but still require some time out of commission.
Then you've got to have training periods - and in the case of Trident they have to wander over to the US every so often for test-firings and to swap old missiles for new. We no longer do maintenance in Blighty, as we did with Polaris, because the missiles have a longer "shelf life" - so it was easier not to bother - minor maintenance can be done from within the sub.
So to have one boat permanently on patrol, you've probably got one in long-term maintenance of some sort or another, one working-up and training to take over and the last one recovering from its last trip.
That leaves some margin of error for accidents, such as a major breakdown or crashing into another sub - or the land. Where you'll have to sortie the boat on work-ups and then hurry the other one out of short-term maintenance - to make sure you're covered. Supposedly we've never not had a boat on patrol (like the MoD would tell us), but 4 gives a tight margin of error if two boats were to have unexpected maintenance issues simultaneously.
All this is brilliantly covered in Peter Hennessy's 'The Silent Deep' - which is about the RN's Cold War submarine program. He's a constitutional historian who got into the Cold War and nuclear policy via that route, and I'd also highly recommend his 'The Secret State' - about the early Cold War and British government reaction to it. His books are proper academic histories, so can get a bit heavy at times - especially when you get into chapters on nuclear policy debates, but he's a witty writer when he can be, and likes to get in a good story or two as well as the dry analysis of cabinet papers.
'Does it REALLY take a year to do maintenance on a sub that has been out 'in the field' for a year?
It's a large lump of metal that gets deliberately submerged in sea water for most of its working life.
Try running your car continuously for a year and see how that works out without maintenance.
To be fair, the innards don't see much sea water.
But the innards of a sub make even the most tightly-packed car engine bay seem spacious. You always have to remove something else to get to anything and there's nowhere to put that something else so it has to go ashore. And if it's bigger than a manhole cover then you need to cut a hole in the hull.
"And even if the operation:maintenance ratio is close to 1:1 then surely with 2 subs you're almost covered and 3 should be fully covered?"
I think some of these points have been covered, but I'll reiterate some.
Firstly, a warship spends maybe half it's time at sea getting ready for active duty, as compared to being on duty. Thus active operation to not active is at best 1:2. Using the same reasoning as above, you can "just" be covered by 3, and 4 is "fully" covered.
This assumes that the maintenance/refit schedule covers all problems, there is never an unexpected issue, and all work is done on time. Since these are all likely to be wrong, it becomes probable that two or more subs might need to be in dock at the same time.
Hence why we arrive at the numbers given by the sardines, 5 for minimum full coverage, 4 should be fine for short periods (ie rearming trips to the USA).
That's odd. I thought the whole point of a nuclear sub was that it could stay submerged for weeks / months on end (basically as long as food/drink supplies last). Does it REALLY take a year to do maintenance on a sub
===
the maintenance i just ONE thing, once the ship has been away from sea for any length of time the crew need to come back up to speed and require training in ALL aspects of their duties before they are deemed fit to return, hence if you have four, one at sea, one heading back, one in maintenance, one in training is how it all runs
It depends, there is regular maintenance which is done pier-side when the crew is switched and supplies are reloaded etc. That's why most navy's with ICMB equipped nuclear submarines have 2 complete crews per sub. In the US they are called blue and gold, so when blue is out on deterrence patrol, gold is training, and integrating new personnel. Then there are more complex overhauls after 10 or 15 years, designed to introduce new technology, or refuel the nuclear reactor. Those can take up a year, including a shakedown cruise to see if all the new stuf works as promised etc.
Are subs not better at this than surface vessels, especially given the "afternoon effect"?
Not if you want positioning to an accuracy of 20cm. I suppose a sub could stick a long antennae up top to break the surface and get GPS, though that only works if your surface region is fairly shallow - not hundreds of metres. But actually the afternoon effect remains of interest anyway - because even if your survey vessels were subs, you still have a significant surface fleet who want to account for it.
For instance, temperature is not the predominant driving factor in sea water density - salinity is. If you're up north somewhere then in the summer, relatively fresh ice melt will (despite being very cold) sit on top of the sightly warmer, but denser saline.
In various parts of the world you get various Pycnoclines of various thicknesses at various depths for a number of reasons (e.g. anywhere you have a significant estuary/river outlet - the outflow will sit on the surface as it is basically fresh whilst the deep water is salty with limited vertical mixing). You do also get vertical mixing/downwelling at the poles ("Deep Water Formation") where the thermal influence eventually dominates the salinity.
It is useful for the Navy and UKHO to profile those gradients so that (for instance) a minesweeper can account for them. Submarines can also carry the full profile and just disregard whatever is above them (unless they're looking for ships or other subs, in which case it may be useful to know what is between them and the surface!).
The one particular place subs would be useful is that - as the author notes, deeper water gives you a wider spread on your survey - but this comes at the cost of reduced resolution. Generally if you want better resolution you deploy a towed fish down to a sensible depth above the surface. You'll get a much narrower survey swath but higher-res data. This is also where subs do make an appearance - you can fire off an Autosub to go do an autonomous survey and meet you again at a designated waypoint. It can gather hi-res data from deep down, but to the cost of lateral positioning accuracy (though inertial positioning isn't bad, but not up to DGPS standards).
The answer is both yes and no. Depends what you are doing and how you are operating the AUV. If you are using something like a Hugen it uses INS and can't get a fix by it's self. So the data is accurate locally but not globally as in if you find two "things" you know very accurately where they are in relation to each other, but less accurately the the datum. If you are operating it directly then you can feed it positional updates via HiPAP (This is a very simplified answer) and you have a much more accurate system but you loose the capability to let it run by it's self. The biggest error in deep water is the Sound Velocity, an accurate SV profile is far more important that getting your GPS 5cm better. In deep water a carp SV can put you bathy out by tens of meters or more.
submarine positions are classified, including where they've been. makes it hard to transfer it to sea charts that have public access.
what makes more sense are robot subs like the ones that found Titanic.
Also worth mentioning: as you go deeper, and pressure increases, that also affects the speed of sound in water. So submerged sonar nearly always travels in a curve... and the sonar scanning result COULD be 'ghosts' like old-style TV antennas often got.
When you're submerged in deep water, surface sonar pings sound like you're inside of a giant cave. It'd actually be pretty awesome for recording music with how 'large' it sounds. Yeah I'd heard a few of them in my day... [and what's funny, if I whistle 'like that' with my eyes closed in a room that echoes well, it's like I can 'feel' where the walls are - the human brain can process it]
I would imagine that one situation where subs would be better than ships would be mapping the seabed under ice.
It might make sense to use a cheaper non-stealth unarmed mapping sub for that. Possibly even a drone sub?
You've actually hit on one of Autosub's very specific party pieces - they have used Autosub precisely mapping not only the seabed, but also pinging "up" and mapping the underside of the ice, which is useful in understanding how sea ice forms..
Autosub 2 is currently thought to be in long-term cryogenic storage - it never made it back to the recovery waypoint from an under-ice mission in 2005.
This post has been deleted by its author
Meanwhile the rest of the fleet are in port with breakdowns, at sea without weapons, or awaiting the outcome of the endless haggling by HMT to try and sell them to whoever will buy them. None of which really matters since they can't recruit enough sailors to fully man the sadly depleted fleet they've got.
Meanwhile, the bungling clowns of MoD still have 57,000 civilian staff sitting on their arses pushing paper.
"Meanwhile, the bungling clowns of MoD still have 57,000 civilian staff sitting on their arses pushing paper.".
Oh dear, what would you like them to do, be involved in some war still be pushing paper.
It's like with the two new floating airfields, long may they parade the Atlantic and the Pacific with the Americans and stay out of harms way, it's a beautiful sight, and the officers are well dressed like in the best opera.
Sound and light are funny, in a way. Light travels fastest when there is no resistance but slower through air and even slower through water.
Sound again is just the opposite, from "nobody can hear you scream in space" to a bit better/faster in air, and a lot faster in metal.
Lay your left ear on a rail trail, send your buddy some hundred meters away with a sledge hammer to hit the rail. You will see him hit the rail and then much later you will hear it through your left ear and then a bit later through your right ear.
And you will never feel the speed of sound is much to boost about.
Then for a beer, if not hit by a train.
Lars,
It's Friday, so just a gentle little point - light travels at only one speed. It doesn't vary according to the medium.
Think of it this way: a beam of light travels from point A towards point B. It's going at the speed of light - it can do no other. Later, a beam of light, also traveling at the speed of light, arrives at point B.
When the time is measured, it seems to have taken too long, and therefore was 'slowed'. It wasn't - it can't be. It took the shortest possible route, as light always does. But, roughly speaking, the route was longer than imagined. Off for a beer now.
@DamnedIfIKnow
No, you are so wrong and why did you not check you facts before writing.
https://en.wikipedia.org/wiki/Light
"The speed of light in a vacuum is defined to be exactly 299,792,458 m/s (approx. 186,282 miles per second). The fixed value of the speed of light in SI units results from the fact that the metre is now defined in terms of the speed of light. All forms of electromagnetic radiation move at exactly this same speed in vacuum."
"The effective velocity of light in various transparent substances containing ordinary matter, is less than in vacuum. For example, the speed of light in water is about 3/4 of that in vacuum."-
Perhaps you just came from the pub writing rubbish like that.
'the layer'
'convergence zones'
'thermocline'
The RN's Submarine Service was exploiting the thermocline during the Great War; during the Dardanelles campaign, it was possible to park one's boat at the boundary, because it floated on the cold water below but sank in the warmer water above, although there is one instance recorded of the crew having done this at night and woken in the morning to find their boat was now on the roof, ie at the surface
How the hell do they do that? "Hellooo, yes you that funny-looking crab, is it OK if we do some surveying"
I would suggest that the funny looking crab is far more worried about his over-enthusiastic yellow sponge of a fry-cook, and his surly, sullen cashier squid to worry about the noise from above. In fact the noise is probably welcome as it will drive the local fish in to the restaurant for some peace and quiet and maybe a burger or two.
Plus he'd just blame it on Plankton anyway.
How the hell do they do that? "Hellooo, yes you that funny-looking crab, is it OK if we do some surveying"
We used to generate the acoustic pings by throwing sticks of dynamite in the water and listening for the echo (there was a lot of surplus going cheap after the war).
This was found to be a bit antisocial and we now use various other methods of creating compression waves (which is all a sonar ping is) that have less potential to upset creatures using echo-location, or indeed concussively stun/kill anything in the immediate vicinity (not necessarily so much of an issue in blue water but considered poor form in coastal surveys!).
They also listen out on the hydrophone and hold off on mapping operations if there are active pods in the area.
Aside from environmentalism, the tech has also moved up that we can use multiple coordinated sources (as mentioned by other commentards) to get a more accurate profile compared with single-source methods. Arbitrarily throwing explosives off the back of the boat isn't accurate enough for those purposes!
The excellent techie podcast Omegatau did a long episode onboard this ship life-and-work-on-hms-enterprise
Surprised that people (including the author) are surprised by this. The RN has been surveying the worlds oceans from its very inception! That is what all the great famous Captains spent most of their time doing on their voyages of discovery. Accurately charting the worlds oceans and sending them back to the Admiralty. With charts come power and with power, wealth and a lot of the "success" of the Empire was built on the back of skilled navigator and survey creating accurate sea chart for others to follow! The Article should really reference his long history of surveying as a RN skill.
The problem with a civilian agency is that you can't be sure who else they'll provide the information to, and you don't want just anyone having the same information your submarines do.
Having said that, the UK Hydrographic Office which produces the actual charts is a money making venture.
hadn't long since been handed over to a civilian agency.
1. The differences between looking for objects and scanning the sea floor if you are using modern multi-beam gear is mostly the math applied, not the actual xmit/rx.
2. You never know whom will you are going spook. Though that is an issue mostly with radar and yanks, not sonar (*).
3. You should not unintentionally disclose the location of your own assets and your schedule must be sync-ed with them. Modern passive listening gear is capable of extracting a hell lot of information out of somebody else making noise. You really do not want to be doing it when one of your subs, drones, etc are in the area.
All of these do not work well with a civilian agency.
As an example of a civilian shining a radar at the wrong thing at the wrong time, this is what happens if you shine it at a trigger happy USAF moron flying a Wild Weasel a few days after someone shoots down an F117. The off-course claimed by NATO PR, is utter ratshit, it was very much on-course for Sofia airport MRL site, just did not take into account the hills in the beginning of the city.
I am not aware of similar sonar incidents, but I would not be surprised if there were some during the cold war. By the way - after the incident the military showed up and put limits on where the radar can be pointed at until end of the Kosovo conflict.
"the point of the United States Army Corps of Engineers"
I always assumed that the point was "We have to have a corps of engineers anyway, because civilian engineers refuse to work in war zones, and we need to keep them trained, so they might as well practice on civilian jobs and save Uncle Sam money.".
Personally, I think that in smaller countries like NZ, they should ditch 90% of the standing military forces and turn them into an engineering/search and rescue/logistics corps. Since that's what they do 95% of the time. Obviously this organisation would be mandated to provide support operations to the actual military if required. Perhaps they'd need small-arms training for deployment to risky zones overseas, but I don't think all staff would need to be "certified" in such a way.
Despite some family urging, I was never interested in joining the military, but if it'd been an engineering corps, I would have joined in a flash.
For the actual remaining military, get them the latest and best gear and training so it's basically like a souped-up SAS team, so they can do anti-terrorism and peacekeeping operations. Because, to be frank, if anyone wanted to invade NZ with all the coastline and 4-5 frigates to "defend" it, they wouldn't have to work very hard.
The reason is in the history. 200+ years ago there were no canals in the US and the roads were crap. So when Congress decided we needed a Military Academy to train officers, they also required the cadets to learn civil engineering. There are other things for cadets to major in now, but the Corps of Engineers is still a big powerful institution doing all sorts of things that have little to do with national defense. (They still build canals and dams, control where you can put bulkheads and seawalls in harbors, and provide the raw material (Potomac water) to the DC Water supply, and I don't know what else.)
The maths is fun
Posting as AC for obv reasons
Back in the day MSc. project (of myself & a colleague, with a company that no longer exists) was on real time distributed computer system processing sonar results from source with multiple sonar emitters used by certain types of boats.
The mathematics (including issues regarding temperature related signal speed differences was "fun", to say the least!) - multiple emitters (and how they were distributed in space) makes interpreting the signal a lot easier, which is why the sonar devices for serious use use multiple sources
Disclosure I do not work on weapons systems, just the vagaries of industry project placement on a higher degree, I'm sure the Masters degree team leaders took delight in assigning people inappropriate projects (myself & colleague were both pacifists) knowing industry project was compulsory element of the degree.
Have been sailing for most of my life, and have used many different paper charts in my time including Imray, NV and others. But I have to say that Admiralty charts and other relevant publications such as the tidal streams and ocean planning atlases always come with a very warm additional feeling of "correctness".
Great article also El Reg.
"Whereas the minehunters are looking for object detection, we're looking at safe navigation."
The difference is mostly in the data transform applied to the received data. You can in fact do both.
It is a matter of what you feed into the Fourier transform and what convolutional filters you apply to the data. While a sonar (or radar for that matter) that looks for objects will never be fantastic for bottom survey, it may be just good enough and vice versa.
The give-away is in the boxes which do the digital processing. They are nearly always from the same vendor and have the same guts. The difference is mostly in the software load.
I did my degree in Navigation\Hydro so did lots of under water acoustic systems and satellite nav stuff. Remember doing a project using side scan to survey the Tamar in and around the Dockyard and having to cut the paper swaths up and stick them together by hand to form a mosaic. Sonar has come a loooonnnnggggg way in the last 30 years!
"I remember doing that for Seismic surveys back in my days of being a Geophysicist - scalpels and sticky tape, high tech stuff!!!"
You might be interested in the latest series of The Curse of Oak Island. They brought a company in to do a near surface seismic survey. Lots and lots of closely spaced, low rated explosions and lots and lots of sensors. The latest episode has only got as far as showing the last of the survey. Hopefully over the next few episodes we'll get to see the results. AIUI, they are surveying for voids from surface to 200' or so depth.
"You mean you actually watch that crap? There is nothing there. There was never anything there. It's just an endless stream of "ooohhh, what a mysterie! What will they find next?""
It's my wife who mainly watches it, but having said that, if you mentally filter out the constant crap theories and the often poor way they go about things, it's still an interesting treasure hunt with the occasional interesting find (even though usually not related to the so-called money pit)
AC,
my [now retired] wife was part of the team that did the same thing to come up with modern [IMRT] cancer radiotherapy treatment.
Part of the research project involved using umpteen scans of a patient, then someone got someone to use a CNC mill to make a metal 'mask' that was used to give a more accurate does of radiation to treat the patient without so much of the side effects of dosing areas that ideally you wouldn't.
Fun stuff eh?
Jay
I just have to ask you if you think it's a coincident that boat is boot in German, båt in Swedish and Norwegian pronounced as in German and of course båd in Danish, boot in Dutch and bateau in French and Bát in Icelandic. But the Klingons did not copy it from English as it's Duj.
Barca and barco in Italian and Spanish.
I'll be impressed when the resolution is high enough to resolve cans of coke and beer on the seabed and plastic bobbing on the surface because Trash>Global Climate Chaos (till it finds a higher state of order which will be less human friendly) is more of a threat to life on earth than any military (Mutually Assured Destruction will never happen).
When I read the sub heading "Warships don't only sink the nation's enemies, you know"
I remembered experiences our navy had. HMAS Melbourne (the last Australian Aircraft carrier) was notorious. Allies were as much at risk as enemies (or more so). From memory HMAS Melbourne did more damage to allied warcraft than it did to enemy.
You may think of a warship as a vessel that sails the Seven Seas, bristling with missiles and guns, ready to deal out death and destruction to Her Majesty's enemies.
If the vessel's name starts with HMS - given recent purchasing/acceptance history of the MoD - then the chances of anything bigger than a .50 cal being operational are pretty low...
He added: "You've got HMS Hood [the First World War era vessel] down there, she was sunk deliberately to block entrances."
HMS Hood was sunk "deliberately" by the Bismarck with a lucky shot that hit a design flaw that allowed Bismarck's Mr 15" high explosive shell to be introduced to Hood's Mr Cordite Magazine.
PO Coleman-Smith didn't say the words in brackets, they were added by El Reg. They're not wrong though, just ambiguous.
HMS Hood was deliberately sunk in Portland Harbour during the First World War, but it's not the same Hood you're thinking of. There have been three ships with that name.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
