Hacker News new | past | comments | ask | show | jobs | submit login
The Mystery of Devil's Kettle Falls (mnn.com)
165 points by oskenso on Sept 4, 2015 | hide | past | favorite | 89 comments



There are mine communication devices that can punch a signal through a few hundred meters of solid rock. Here's a simple one.[1] VLF can get through, with enough power, a simple signal, and a smart receiver.[2]

There's been a lot of work on "down-hole communication" for oil and gas drilling. Modern drilling involves com links to the sharp end, with info coming up and commands going down. Some of the techniques work through the drill string, but some are VLF radios.

There are also high-powered ultrasonic pingers, with a range of several kilometers in water.[3]

If you could get info for the first kilometer or two, you'd at least know where to look for the rest of the path.

[1] http://radiolocation.tripod.com/ [2] http://onlinelibrary.wiley.com/doi/10.1029/2010RS004378/full [3] http://www.sonotronics.com/?page_id=96


But that only would work if the hole leads to an open tunnel. If there's a gravel bed on the bottom, your equipment just sits there.


But at least you get a clue there might be a gravel bed.


For the record lava tubes can happen in flood basalts; I've been to some at Hell's Half Acre in Idaho, and I'm sure there are others. But these are not like karst-type cave systems; they're much shorter.

It's really unlikely that the water goes anywhere other than the lake. It just might travel through a fracture network and seep out of the lake bottom miles away, but groundwater tends to flow broadly following the topographic slope (driven by the gravitational loading of groundwater surface called the 'hydraulic head') and there typically aren't isolated tunnels where some water can flow unimpeded by regional groundwater flow.


I'm reading this and wondering why you couldn't build a waterproof beacon? At least by tracking it you could prove that it didn't exit into Lake Superior.

https://en.wikipedia.org/wiki/Underwater_locator_beacon

http://www.rjeint.com/aviation-underwater-locator-beacons


You should drop synthetic DNA in it and then sample all nearby lakes and do PCR. Even aftera trillion-fold dilution it should be detectable. Would be cheap, easy and very likely to work. A cool weekend project for your kids!


> You should drop synthetic DNA in it and then sample all nearby lakes

Sounds like the backstory for a Lake Placid or Piranha-style movie...


In a typically-HN thread full of armchair experts brainstorming obvious "why don't they try" ideas [1], this is a truly interesting idea that I bet the experts haven't considered. Also some good potential in the area of espionage, if it already isn't being done.

1. See also the Malaysian Airlines disappearance thread.


Having seen this crop up on reddit before, with no novel or technically plausible ideas suggested, I have to say, Hacker News delivers :)


This would not be permitted for any NIH funded lab, there are pretty strict rules about handling recombinant DNA. There might also be laws preventing this, I'm not sure.


Why is this so if the DNA isn't part of an organism?

I guess there is some risk in some circumstances that the DNA will affect organisms just by being present in their environment, without having been actively introduced into their cells?

https://en.wikipedia.org/wiki/Transformation_%28genetics%29


Pardon my ignorance on this - but wouldn't dropping Synthetic DNA into the water be polluting it in some form?


The article mentioned locals claiming to have pushed down a car one night ... No idea what volume of synthetic DNA would be sensible here, but I can imagine it'd mass less than a car.


Almost every living thing contains DNA (the possible exception being some viruses if you consider them living).


I imagine leaves / fish / spit / what have you fall into the water all the time.


The only synthetic part of it is the sequence - on a molecular level it's exactly the same as naturally-occurring DNA.


There's a good chance a beacon would get stuck or lodged somewhere, even though the water itself would continue to pass through (think sand or very narrow fractures). Hydrogeologists use a variety of dyes or other water additives that are better suited for this, and it doesn't seem to have worked yet.


>> Hydrogeologists use a variety of dyes or other water additives that are better suited for this, and it doesn't seem to have worked yet.

The one thing the article failed to mentioned is the possibility they have no idea where to look for the dye to be coming out. It's easy to do this on an island (such as the famous Oak Island), not so easy at Kettle Falls where this waterfall is quite a ways inland and the only logical outlet would either be the Brule River, or a deep cavern that leads to the coastline of Lake Superior.

Add in the fact in some areas the lake can be as deep as 1,300 feet, and you've got one hell of a hard time trying to find out where the water goes.


I'd place beacons in live fish. They are well suited to navigating underwater hazards.

Acoustically ping a serial number for a couple of days. Place an array of sonobuoys in the likely outlet area, scan with boats further afield. If there is a fish sized outlet, then they should be able to get out. If you start finding your fish you can start refining the resolution of the search and plopping more fish in the falls.

At the same time, lower teathered hydrophones into the falls at various distances. If the pingers accumulate somewhere there, then you have a suggesting of fish impermeable structure.

I might need to add lights to the fish so they can see.

Plus you'd get some great YouTube footage a fisherman who catches a fish with headlights.


Use something with as close to exactly the density of water as you can manage.

Put a 1 week timer on it that releases some air inside it (a chemical reaction would be easier than compressed air) to make it surface.

Make a bunch of them, in case one gets stuck anyway.


Find an organism that passes through the passage and a way to track that.

Hitch your wagon to something living that will be able to find a way.


A fish with a GPS tracker perhaps? If they attach them to ground animals, why not water?


They also/already attach them to whales, sharks and dolphins, at the least. I bet someone, somewhere, have done it to a fish...


The limit is probably the size of the tracker. The ones I saw look about as big as a standard river fish..


Try otters. They'll eat the zebra mussels when they're done.


You'd need something super specialized, because water and rocks/earth block radio waves exquisitely well, and the churning water would be difficult to survive regardless. So perhaps some sort of camera + high sensitivity inertial measurement unit in an extremely rugged housing along with an ultrasonic data transmission system or something. All of which would be extremely custom and would likely require several iterations and vast expenditures of money and effort just to collect a little data.


Not that you'd find it again unless you attached it to a piece of string.


I'm just assuming it'd be gone forever, which is why you'd need some sort of wireless data transfer system (ultrasonic sound, for example).


Small GPS receiver card (chip?), GSM module with prepaid SIM, a bit of flash storage to record GPS data and a battery - all sealed in a rugged ball. This shouldn't be that costly, no? Or will it be hard to make the battery last beyond few days?

At the very least this should give the first surface point for that undergraund stream.


Water is essentially non permeable by radio waves at the frequencies you would use for GSM and GPS, which is why communicating with submarines is extremely tricky. I've always assumed (but never been able to find information on) convicts with GPS tracking anklets taking baths and unknowingly getting in trouble for breaking the recorded path.


Both the GPS and the phone module is going to be useless while underground, so perhaps it will be better to turn them off altogether and use a separate timer to reconnect them after a few days. A disconnected battery can last quite a long time.


Even an energized battery, on a low duty cycle, (Transmitting once a day at low power), can last upwards of 20 years.


You'd probably need to make it very slightly more dense than water, or it may get trapped at the surface in an underground cave. This means it won't surface by itself once it leaves the cave system, and as water blocks RF pretty effectively, you'd not get a signal. To make it work, you'd need to add some additional mechanism to allow it to become more buoyant over time - perhaps a slow chemical reaction creating gas?


The earth is really good at blocking radio waves; the beacon is only going to be good when it shows up again in water open to the sky.

Similarly, it may be that the water is getting to Lake Superior just fine, but that the path it takes has plenty of opportunity to get debris caught up or lost in a side-channel.


A ping pong ball or dye isn't going to work if it is surfacing in Lake Superior. That lake is so vast that the kind of a tracker they use in airplane black boxes is a good bet. Short of mapping the lake bottom, which is unrealistic, we may never solve the mystery.


The US Navy has been quietly mapping ocean bottoms for decades, since nuclear submarines started going all over the oceans.[1] The resolution isn't all that great, but by now they know where most of the seamounts are. It's a boring job. Cruise across ocean in a straight line towing a sonar array, return along a parallel track, repeat.

But maybe nobody has bothered to do the Great Lakes in detail yet. Not much submarine traffic.

[1] http://oceanexplorer.noaa.gov/explorations/lewis_clark01/bac...


> Short of mapping the lake bottom, which is unrealistic

Is it? http://www.telegraph.co.uk/news/earth/environment/archaeolog...

Granted, these guys seem to be piggybacking on massive work done by oil companies, but the North Sea is roughly 3x Lake Superior, so you only need 1/3 of the effort. You just have to find an excuse...


A radioisotope might work, though good luck convincing people to dump tonnes of radioactive marker into a hole that nobody knows where it goes. A pinger from an aeroplane would do an alright job, but they are usually specified to work for a week which might be far longer than the period underwater.


Exactly, something with intermittent GPS pinging so that should it ever resurface we get a signal.


I'm pretty sure you're not going to get a GPS signal from satellites wherever that water is going.


And what happens if somewhere down below the temperature suddenly rises and all those plastic beacons with some soldered electronics inside simply melt or break?


It there was a lot of heat down there we would know - there would be hot springs somewhere nearby.

For plastic to melt it would be hot enough to make steam, and we would see that.


Wow. So many armchair geologists, both here and in the comments on TFA. Do you really think people haven't already thought of a dozen reasons why GPS beacons/synthetic DNA/blah blah etc wouldn't work?

Should we get some geologists in to comment on your issue tracker and propose solutions for your bugs?

Ninja edit: https://xkcd.com/793/ applies to programmers, too


Alternately, it can be entertaining and mentally stimulating to brainstorm ideas and then pick apart why they wouldn't work.

It would be one thing if it was a funding pitch based on uninformed proposals but as a forum conversation on a Friday afternoon, it doesn't seem too awful.


Wow, is the snarkiness really necessary? At least by proposing a possible solution here, we increase the chances that a bona fide geologist will come by and tell us why it wouldn't work.


> Do you really think people haven't already thought of a dozen reasons why GPS beacons/synthetic DNA/blah blah etc wouldn't work?

Yes.

> Should we get some geologists in to comment on your issue tracker and propose solutions for your bugs?

Yes. And if they're wrong? So what? Is that bad?


See also "I don't know anything about"

https://youtube.com/watch?v=GlcuV_Dojwg


On the other hand, some of the best ideas in any field can come from outside that field. There is benefit to not being constrained by conventional wisdom (which often times can be outdated or just wrong).


Use a long thin cable on a motorized spool. Rig up a waterproof reinforced box with sensors that detect speed and direction (mercury switch type deals?... I don't know but I bet there is something). Hook them up to a board with some storage. Drop the box in and start reeling the cable off. If the cable goes slack the thing has probably stopped. Maybe it's hung up. Pull it back a bit and try again. If it keeps stopping in the same place it's probably all the information possible by this method. Retrieve the box, get the speed/direction information, do some distance calculations and graph it. Ideally the box will go all the way to the exit point and you will have a profile showing where the water ends up. But even if it doesn't, you will have a good idea of direction and depth the water is traveling for a certain distance at least. This will at least make better guesses in the future.


I've boated the Brule. That is a definite portage ;)

Better photos here: https://roadtrippers.com/stories/revealing-the-mystery-behin...


They don't have to dam the whole thing but they could redirect the water going in the kettle to go over the falls on the right, so the hole is free to be explored via winch & harness, or remote camera.


Why don't we block it off for a few months and see if it dries up?


Lake Superior?


No, I believe the idea is that if water wasn't flowing through the pathway that we would have a lot more options for exploring where it goes with various equipment.


Has anyone attempted to map the first few hundred meters of the kettle? That might give us a good idea of where to look. Drop a sonar on a very long, strong wire to map out the contours?


It could just go down into an underground aquifer, no?


Sounds like a great place to dump a body...


I know right? If that hole could talk...


Dump radioactive waste in, see where the Geiger counter picks it up?


Maybe they could use radar sounding and a big bag of passive reflectors floating upright.


are there any short life radioactive materials that can be used?


Iodine-123 maybe, but you'd have to make it on the spot.

The half-life is 13.22 hours, and it makes some detectable gamma rays.


Why not lower a camera equipped robot with a wire there?


How will you stabilize it to get a decent picture?


why not just cut it open and see where it goes?


Last time this was posted, I suggested dropping a nuke with a 12-hour timer down the hole. I got downvoted too. :(


I thought of that, too, but there is no need to be able to detect it from the other side of the earth. Timed depth charges and sensitive seismometers should be enough.

More ping pong balls also should work. The first few billion may get stuck along the ceiling of the cave, but if you drop enough, some will have to come out somewhere.


There's a very simple explanation for why anything that floats wouldn't ever appear in the lake, even though the Kettle drains to it:

    | | Kettle
    | |
    |~|    |~~~~ Lake
    | |____|
    |___________
The water level would equalize between the Kettle and the Lake, but there's no way for anything that floats to get over.


That's only semi true. The rushing water would carry even things that float.

If it didn't the entire hole would be full of branches and trees, but it's not.


Only if the flow rate is high enough to defeat the buoyancy. Not guaranteed at all.

Also, there was no indication what the geography is within the hole. Organic material will eventually break down and waterlog enough to lose its buoyancy. Those ping pong balls and rumored car, though...


Well, cutting open is perhaps a bit drastic, but how about just stopping the river for a day or so? After all, the OP mentions that this hole appears at the split of a river, so diverting the water to the other arm of the river might not be that hard...


I've learned years ago that even the "interesting" things have little chance to happen if they cost money. While visiting one of the most interesting caves in Poland the guide said that geologists strongly suspect that there's an entire tunnel system underneath it, and have suspected it for 40-odd years. So I asked - well, it should be fairly easy to check with the right equipment, no? And he said that yeah, it should be fairly easy, and even though multiple universities expressed interest in doing it, no one is willing to cough up the money for it. I guess this is very similar - of course it would be extremely interesting to find out where that water is going,but redirecting the flow of water would be very costly, and apart from satiating some scientific curiosity there's little reason to undertake such large engineering project.


Just tell them you think there is a train full of Nazi gold in those tunnels :-). Something we don't have enough of these days are "gentlemen explorers" folks who are independently wealthy and spending their days exploring interesting places.


Bear Gryls?


I had a similar experience here in a shallow cave in Texas. When you reach the 'end' you really just reach the point at which the cave is flooded. The cave has been well known for over a hundred years and likely goes on quite a ways. No one really has an interest in exploring it. It'd be an expensive and likely just discover a bunch of rock and water.

With respect to cave systems, it seems like that the key is finding a previously unknown species. Or find a species in the cave that was thought to have a very limited distribution.


Not to mention very dangerous, cave diving is a notoriously deadly enterprise.


I bet you could do something neat with small tethered submersibles that you remotely control. Of course, they might not be small enough, but at least it would be safer.



This sounds like a really interesting company.


The water is just going to stop at the level of the lake, so it will still be flooded. By far the most likely answer is that it gets filtered out into the bottom of the nearby lakebed. I'm guessing there are pockets above the passage that the water flows through where floating debris gets trapped (full of ping pong balls, etc...) and the flow rate is too high for dye to remain detectable.

Damming off half of the river and then cave diving it would be awesome, but tremendously expensive and only good for answering a trivia question.


...where would you put a day's worth of river water?

Random internet search shows that a small river flows at about 1000 cubic feet per second, which is about 30 cubic metres; that's three million cubic metres in a day, which is three million tonnes (water is so convenient). The biggest supertankers can carry about 300,000 cubic metres, so you'd need ten of them...


"...where would you put a day's worth of river water?"

The other side of the river, by diverting all the water down the falls. River flows certainly vary by more than half in this part of the country/world (I'm in Michigan but it's not so different), so if you do it during a down time of the year you would not be creating a flow the rest of the river bed could not handle.

I'm not saying this is a "good" idea but it isn't a bad one in the sense you're suggesting. There is a place to put the water.


Wait until the dry season so you have the lowest total volume, then send it down the other fork, treat it like a flood?


Maybe people live downstream?


Wait until they actually try this, and then nothing happens :)


dude people are just mad. but the purpose of this account is to post what i want, and try not to be effected by the down votes.

at the same time i just wanna see it go to 10k


Ground penetrating radar?


Does not work very well in a wet environment (the energy is absorbed pretty fast). And if you use longer wavelength to get better penetration the spatial resolution goes down hard. It might help to map the part above though..

(No idea for longer wavelenghts, but we use GPR to measure ground water content in the MHz range. Normal penetration depth is at most a few meters.)


Wow, that looks dangerous as hell!




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: