Hacker News Reader

Comment by wolverine876 | Hacker News Reader

The mountain is in the South China Sea, a leading focus of the US Navy. We learn something about the state of underwater mapping from the fact that even in the South China Sea a mountain wasn't mapped.

Why can't satellites map the topology using something like RADAR (though likely using other frequencies), which I believe is done on land? Is there no technology that will penetrate water sufficiently? Is it a physical limit of electromagnetic energy and water?

I assume it hasn't been done by satellite because otherwise we wouldn't have unmapped mountains, and because the same technology could help find submarines, taking away their stealth and eliminating a nuclear deterrent.

View on HN

Civilian geophysicist (but not an oceanographic one) here. This is my best understanding:

As others have posted, above surface based propagating EM techniques won't work because seawater is conductive.

Acoustic waves are probably the best way to map bathymetry, but that requires active sonar. The subs don't use that because it gives away their position. Surface mapping (sidescan sonar) in the area is probably the best viable technique to do the job, but geopolitical concerns make it impractical.

Aside from sonar, gravity and magnetostatics could in principle be used to find seamounts. But both suffer from a fundamental problem that "upward continuation" decays short wavelengths exponentially faster than long wavelengths. In other words, to horizontally resolve a seamount with a characteristic length of (say) 10km you'd need to fly no higher than 10km. That rules out satellites. What's left are airborne and surface based surveys, which face the same geopolitical issues as sidescan sonar.

Give all of that, you might as well do the sonar survey, as long as you can solve the geopolitical problems.

Gracias.

> above surface based propagating EM techniques won't work because seawater is conductive

Dumb question: My very poor understanding is that the EM images are created by objects reflecting the energy back to the sensor. Is the problem that the water's conductivity will spread the EM in every direction or otherwise cause it to behave unpredictably?

> "upward continuation"

It's mostly that the inbound energy doesn't even reach the reflector. An optical analog (yes, yes, EM true) would be trying to image something well behind a mirror.

"Upward continuation": both gravity and (after certain manipulations) magnetostatic surveys obey Poisson's equation. The characteristic (Green's) function has a 1/R decay. If you will allow me another optical analog, upward continuation is closely analogous to a detailed image being convolved with a 1/R blurring function. That's the heart of why I (roughly) say you can only resolve stuff of characteristic length R by observing/measuring from below a height of R. Higher than R, and the short wavelengths are completely attenuated, and you lose ability to resolve them.

Thank you; that makes some sense!

There are no geopolitical concerns which prevent the US Navy from doing side scan sonar bathymetry in the international waters portion of the South China Sea. In fact they do this on a regular basis. But they haven't covered the whole region yet.

Which does kind of beg the question, why did a fast attack sub hit an uncharted seamount? (If that is indeed what happened. Maybe it was "user error" in which case it sucks to be the sub's captain and sonar operator.)

> There are no geopolitical concerns which prevent the US Navy from doing side scan sonar bathymetry in the international waters portion of the South China Sea. In fact they do this on a regular basis.

China, a powerful country, says it's their territory and threatens those who disregard it. Many think it's the most likely place for a US-China war, besides Taiwan. That's a geopolitical concern.

The US chooses to handle it in a certain way, including by operating in the area. The US says that legally there is no concern. But that doesn't preclude a geopolitical concern.

Radar does not penetrate through brine water. Other remote sensing techniques have such poor resolution that they are relatively useless for this task. Unfortunately, sonar based bathymetry is really the best tool for this task and that means you have to run around with a surface ship mapping out the sea. As you can imagine the Chinese are not a big fan of American ships doing that in waters that they claim to be theirs.

> sonar based bathymetry is really the best tool for this task and that means you have to run around with a surface ship

Why can't it be done from a plane or satellite? I can imagine answers (e.g., sound doesn't travel as well through air), but does anyone know?

> the Chinese are not a big fan of American ships doing that in waters that they claim to be theirs

I'd be surprised if that stopped the US in this circumstance. They regularly conduct operations in the South China Sea.

Sound doesn't transmit well at all between air and water in either direction. It just bounces off. You can't hear what's going on underwater when you're out of the water and you can't hear what's going on above water when you're underwater.

How do anti-glare coatings on lenses work? Could something similar be devised for the air-water interface?

Sure. You just need to apply your anti-glare coating to the world's oceans.

I mean for acoustic waves. Alternatively/similarly, could an aerial sonar waveform be matched to the ocean surface waves to cause constructive interference below the surface, and destructive above it? Would it be possible to recover the reflections afterward?

https://en.wikipedia.org/wiki/Anti-reflective_coating

> Layer thicknesses are chosen to produce destructive interference in the beams reflected from the interfaces, and constructive interference in the corresponding transmitted beams. This makes the structure's performance change with wavelength and incident angle, so that color effects often appear at oblique angles.

Water isn’t air? Even air can become opaque.

Sound doesn't travel as well through the air, but it still does. The primary reason that you can't do it from the air is that the surface of the ocean is a big irregular reflector of acoustic energy from the air, so the overwhelming majority of the energy will be lost. It just would never be practical.

It might be possible to use airborne gravity gradiometry to map seaborne mounts. I'd never considered that before, but I see no reason to think it wouldn't work.

I'm finding it hard to believe that satellite and/or local gravimetry would entirely miss an underwater mountain.

The kind of sensors you can put on a plane or satellite simply can’t penetrate seawater effectively. It absorbs radar, dissipates heat and light. Sonar sound waves in air mostly just bounce off the ocean surface.

This is why submarines are so stealthy and therefore militarily useful. As long as they stay quiet they’re almost impossible to detect without using relatively short range active sonar.

The air / water interface makes it impossible to get any useful sonar data from an aircraft. Some military helicopters carry dipping sonar that they can temporarily lower into the water, and specialized aircraft can also drop sonar buoys, but those are designed for detecting submarines and are mostly useless for bathymetry.

The US Navy does regularly operate in the South China Sea. In fact the USNS Mary Spears oceanography research vessel was there recently, presumably gathering bathymetry and other data.

https://www.marinevesseltraffic.com/vessels/USNS-Mary-Sears-...

Thanks.

(I'd upvote you but for some reason votes don't stick, on comments and stories: When I reload the page, the arrows return, as if I hadn't voted. flag/vouch behaves normally. Perhaps a mod will read this.)

Votes typically take a few seconds to resolve, possibly longer under load. HN's JS indicates them immediately when you're on the page.

Check your profile's upvoted comments / posts link (https://news.ycombinator.com/upvoted?id=wolverine876&comment... -- that's private for you only) to see if your upvotes have been registered.

If not, drop a line to hn@ycombinator.com as I suspect they'll be interested.

Thank you. No upvotes in 2 days via that link and I'm not that stingy! I upvoted your comment just before I wrote this.

I'm confirming either no upvotes (or at least balanced ones) on my GP comment above.

What browser are you using? We changed hn.js a few days ago to use the Fetch API - I wonder if this is related.

It's probably not a configuration you test against (and almost certainly not worth testing against), both an unusual and an old browser used for compatibility with something else.

  Tor Browser 8.5.5
  - built on Firefox 60.9.0 ESR
  NoScript 11.2.11
  - ycombinator.com is 'trusted', which has always
    been sufficient for voting to work

So I'm not filing a bug report, so to speak, but if it helps you track down something useful:

* It happened once before, I think sometime in 2020 (sorry I don't recall more precisely), and it likely would have been in the same browser. I forget what fixed it but it went away eventually. Possibly it was just restarting the browser, which I can try if that will help (but that will lose the current state, of course, and Tor Browser deletes a lot of the current state on a restart; it may create a new profile).

* The top comment on my upvote page is this one from 2 days ago: https://news.ycombinator.com/item?id=29062312

* I don't know the precise times of my comments, of course, but if you see I was commenting then I was very likely voting (though my voting slowed a bit in the last day or two when I noticed they weren't registering).

* I just upvoted your comment at 22:44:07 UTC

I think the problem is that our JS now does this:

  fetch(url);

where it used to do this:

  new Image().src = url;

... and that seems to breaking a few edge cases of older browsers.

Can you try collapsing a comment somewhere and then refreshing the page to see if it stays collapsed? If it doesn't stay collapsed, that's another sign that fetch() calls aren't making it to the server.

> Can you try collapsing a comment somewhere and then refreshing the page to see if it stays collapsed?

I tried a few times on a couple pages, and they don't stay collapsed after refresh.

EDIT: Using another browser on this computer, I just logged into HN, and upvotes and collapsed threads seem to stick. I upvoted your comment (i.e., the parent to this one I'm writing) at 00:27:49 UTC.

This browser (another old and unusual one, but much more recent):

   Brave Version 1.25.72
   - built on: Chromium: 91.0.4472.101

It looks like you managed to upvote https://news.ycombinator.com/item?id=29101929 - did you do anything different to get that to work? e.g. different browser or turn JS off?

Yes, different browser. See my edit in the GP (which seems to have crossed on the wire with your comment).

I reverted the change for voting and collapsing. Does it work now?

Yup, both upvoting and collapsing work. Thanks! Though I wouldn't want to be the edge case that stands in the way of an update. I can workaround it pretty easily.

Another user had emailed to complain of something similar. Where there are two there are probably many, so I think we can leave it this way—it's no big deal. One question though: do 'hide' links on stories work, e.g. on the front page or /newest? I suspect they probably won't, because I didn't revert that part of the change.

> One question though: do 'hide' links on stories work, e.g. on the front page or /newest? I suspect they probably won't, because I didn't revert that part of the change.

I've never used that feature so I don't know what's supposed to happen ...

1. On /newest, under a story, click "hide"

2. Story disappears, stories beneath it are renumbered, an additional story appears at the bottom of the list.

3. Reload page (F5)

4. Hidden story reappears

EDIT: Also, I just noticed that when I upvote a story, the points displayed under the story don't increment until I reload. For all I know it's always been that way, but at the moment, of course, I'm a bit more attentive. Not a problem for me, but maybe useful for you.

> As you can imagine the Chinese are not a big fan of American ships doing that in waters that they claim to be theirs.

Why does anyone (besides China) care? Nobody in that region recognizes those claims.

> Why does anyone (besides China) care? Nobody in that region recognizes those claims.

China has power, military and economic, and thus can't be ignored. If Fiji claimed a region of the Pacific, it wouldn't matter (in this respect).

>> Why does anyone (besides China) care? Nobody in that region recognizes those claims.

> China has power, military and economic, and thus can't be ignored. If Fiji claimed a region of the Pacific, it wouldn't matter (in this respect).

It isn't ignored, but that doesn't mean those claims are recognized. The US (at least) takes actions to specifically show that:

https://en.wikipedia.org/wiki/Freedom_of_navigation#Freedom_...

> It isn't ignored, but that doesn't mean those claims are recognized.

Agreed. The GGP was asking if anyone cared. People certainly care about the claims, even if they don't recognize them.

Because China happens to have nukes, probably.

A good techno-thriller might be some group of hackers nab the Chinese bathymetry database, which then finds its hands in some war hungry US general

> Why can't satellites map the topology using something like RADAR

If you were close enough, say in a plane, you could do something to cause a soundwave to propagate through the water. Say you fire a femtosecond laser pulse or blast it with an ultrasonic pulse. You could then, probably, use a second laser to pick up the reflections as disturbances on the surface. Otherwise you're probably stuck, since RF doesn't propagate well through a conductive fluid. Submarines can communicate somewhat using very, very low frequencies of electromagnetic energy, but the problem with that is the resolution at those wavelengths would be very, very poor.

The best bet at this point in time is probably to just make a bio-mimicing, underwater drone and map under the surface. If you were really clever, you'd just hook some electrodes into a whale's brain and make it carry your mapping payload, using whale sounds to map the ocean floor. Not very nice thing to do to a living creature, but we've done very bad things in the name of national security before.

The best bet at this point is to just tow a powerful side scan sonar behind a surface ship, as has been done successfully for decades. No need for any overly complex Rube Goldberg schemes.

I'm talking about areas where you need to covertly surveil the seafloor(say, in waters near China). Of course you wouldn't need any of those schemes for a normal mapping operation.

Yes, water is a very excellent RF attenuator at all frequencies. Acoustic waves travel quite far in water, especially in the ~1 kHz range. Making loud sounds isn't something you want to do when you're somewhere you're not supposed to be though. If a mountain isn't making any noise then it's going to be difficult to know that it's there.

I think it’s not just possible but done[1], I do wonder why it still happened though. Maybe maps were outdated or resolution is not ideal?

1: https://www.esa.int/ESA_Multimedia/Images/2012/05/Mapping_th...

That radar is actually measuring the sea height. Sea height (on average) more-or-less defines the geoid (the height of an equipotential surface) over the oceans. That ignores sea height disturbances due to winds, currents, tides, etc. The geoid, in turn, is sensitive to local mass distributions. So yes, "it has been done", but the result is nowhere near the resolution or interpretability of a sonar survey. Still, it would greatly surprise me if the military _wasn't_ already using the technique. So, if there were interpretable results from the technique, I would personally expect the Navy already had them...

Electromagnetic energy doesn't penetrate water, at least not enough of the right wavelengths to be useful for mapping. The US Navy and NOAA do sea floor mapping using sonar. But it's a big ocean and some areas haven't been covered very well.

Why is SCS US’s business? No I’m not from 50cent army.

Generally the U.S. leads a coalition of western nations that protect international shipping via freedom of navigation exercises. So if some country like India declares it owns the Indian Ocean, or China declares it owns the SCS, or Egypt announces it owns the Mediterranean, then the U.S. (and other nations like the U.K., France, etc) will sail through the area as well as acommpany shipping traffic to demonstrate that, no, these are really international waters that anyone has a right to use, and that no one can claim for themselves.

This is an important and useful exercise in order to keep important shipping lanes free. This is critical to the functioning of the global economy, on which the U.S., U.K., and other nations depend. It is in their interest to ensure that no nation claims ownership of these global transit routes.

Great answer, and I think there are much broader reasons too:

International order is fundamentally anarchy. There is no power - no world government - that can enforce order (the United Nations is just an association, a place where sovereign countries can meet and make arrangements with each other). If the order breaks down, if 'might makes right', then international wars start; wars are bad or catastrophic for everyone. WWII was so destructive that the victors worried that the world could not survive another one (one reason they started the UN and what became the EU) - and that was 75 years ago. The 'US-led rules-based international order' strives maintain an order based on rules (laws), not on might.

(Yes, this is all very simplified; motives and actions are not so simple.)

A simple example is when Iraq invaded Kuawait in the early 1990s. They had the might, so they took it. The US, openly stating they would maintain international order, drove them out of Kuwait. To reinforce the point, they did it according to international law, and with a coalition of international partners.

China seizing the South China Sea is a serious threat to that order - what will China seize next? If the only way to stop them is to fight a war - if there is no rules-based way to do it - the world and the people in it are in big trouble. So the US and allies (which is not only western, but democracies all over including Japan, India, etc.) work to undermine that claim.

Reply on news.ycombinator.com