Wikipedia:Reference desk/Archives/Science/2014 February 24

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

February 24

Whirlpool

During a flood, what conditions (if any) can result in the formation of a whirlpool? In particular, is it plausible for any malfunction at a water treatment plant affected by the flood to cause a whirlpool? Thanks in advance! 24.5.122.13 (talk) 03:21, 24 February 2014 (UTC)

Whirlpools are caused by the geometry of water flow: an influx at the surface that is not co-linear with an efflux at a greater depth. This has nothing to do with water treatment plants per se. μηδείς (talk) 03:46, 24 February 2014 (UTC)

Maybe only if the water broke through the foundation of the treatment plant and opened a hole into an old salt mine shaft... just as an odd example. ←Baseball Bugs ^{What's up, Doc?} carrots→ 03:57, 24 February 2014 (UTC)

Thanks! So what kinds of flow geometry (other than Bugs's example) would favor the creation of a whirlpool? (For example, might one be created if storm surge rapidly flows into a downtown area with narrow streets and tall buildings?) 24.5.122.13 (talk) 04:12, 24 February 2014 (UTC)

As I am fairly certain someone already said, you have to have an inflow and an outflow that are neither in a straight line nor at the same depth. μηδείς (talk) 04:15, 24 February 2014 (UTC)

Gently, there. In general, it would be rare to have whirlpool formation, although eddies may occur, often downstream of an object, where there are large flows of water – swirls that can alternate in direction. This phenomenon depends on flow speed and object size. A proper whirlpool usually will arise as already described, where the flow exit is quite confined, and the water arrives with velocity so that it has angular momentum. The exit point drawing water to it then results in the rotational velocity increasing, forming a whirlpool. This usually is not the situation such as you describe, since the exit points tend to be at the periphery, not in a place where the water can circulate around the exit point(s). It is difficult to imagine the necessary conditions without a sinkhole draining the water, as per Bugs's suggestion. —Quondum 06:18, 24 February 2014 (UTC).

Only the "treatment plant" topside detail is odd about the example, Bugs. Otherwise you're pretty much describing what happened at Lake Peigneur. DMacks (talk) 17:56, 24 February 2014 (UTC)

According to our article whirlpool - the swirl formed as a bathtub drains counts as a whirlpool. That being the case, it seems likely that any large-scale movement of water will contain numerous small whirlpools. So the question is not whether they are present - but the size and ferocity of them. That leaves us with a fuzzy definition and that makes this a tough question to answer. Obviously if we take the literal definition of a whirlpool offered in our article ("A whirlpool is a swirling body of water produced by the meeting of opposing currents.") - then almost anything that creates or bifucates and redirects water current has the potential to cause a whirlpool of some size or another. So, yes - a malfunction at a water treatment plant would almost certainly cause any number of tiny whirlpools...but I'm guessing that you're asking about whirlpools of some particularly large and destructive scale...and it seems to me that all it takes is some large, fast currents going in opposite directions to make that happen. So, yeah, a water treatment plant failure could do that - but so could a perfectly well-functioning water treatment plant...or a bridge or a school or a car dealership or...whatever. Anything that's able to redirect some of the water flow, change it's direction and then allow it to merge back into the main stream from some other direction could be responsible for a reasonably-sized whirlpool. SteveBaker (talk) 14:14, 24 February 2014 (UTC)

Now if you want to talk about sewage treatment plant malfunctions, those could cause water to back up in (and out of) the rainwater sewage system, then, when cleared, the water would flow down the sewer gratings rapidly, possibly causing a small whirlpool over each. StuRat (talk) 17:20, 24 February 2014 (UTC)

FYI, I'm specifically interested in a whirlpool large enough and stable enough to suck in and drown a swimmer -- perhaps also a small powerboat, although the latter is optional. As for StuRat's suggestion: sorry but I can't use it, because in my scenario the storm surge will be coming IN, so any water in the storm sewers will backflow instead of draining. (You can probably guess why I ask -- same reason I've been asking all sorts of questions about crosswind landings, avalanches, whiteouts, hypoxia, etc., etc.  ;-) 24.5.122.13 (talk) 06:31, 26 February 2014 (UTC)

It is not all that uncommon for large underground caverns to be eroded in cities and other places where there are water flows, especially if there is an underground burst pipe that is not diagnosed for years that progressively erodes a lot of rock and soil. Eventually this may cause a surface collapse into sinkhole that can swallow cars at (I've known such to appear in a major intersection in my home city without warning). This could plausibly create a draining point for an inrush of water, and could produce a short-lived (tens of minutes?) whirlpool provided the terrain and flow were suitably aligned, until the underground cavern had filled. Dunno whether this suits your purposes. —Quondum 06:54, 26 February 2014 (UTC)

Maybe it might -- is the geology in southern Florida (Miami area) suitable for formation of such undetected sinkholes? (An additional factor is that the sinkhole has to open during the storm, not before.) 24.5.122.13 (talk) 09:37, 26 February 2014 (UTC)

Such a feature would have to be next to, or within, a large body of water. The typical Florida sinkhole is on dry land, and slowly fills with rainwater until it becomes a pond or lake. ←Baseball Bugs ^{What's up, Doc?} carrots→ 15:30, 26 February 2014 (UTC)

But would it be plausible to have one develop undetected near the shoreline, and then open up without warning when submerged under the storm surge? 24.5.122.13 (talk) 07:10, 27 February 2014 (UTC)

Fibers

Hey guys, Does cooking vegetables with prebiotic fibers like Sweet potato or Jerusalem Artichokes ruin some of their fibers? if yes, how? Thanks ! 109.65.3.190 (talk) 09:55, 24 February 2014 (UTC)

I don't suppose those two vegetables are very different from others - the chemistry of cooking doesn't change significantly whether it's a sweet potato or a carrot. Our article dietary fibre doesn't seem to mention the effects of cooking on water-soluble fibre (the "prebiotic" kind - to pick a trendy name). Other sources suggest that cooking vegetables down to a mushy consistency does reduce the amount of useful fibre - but cooking them to a firm consistency does not. Unfortunately, potato does need to be cooked to a fairly mushy consistency in order to be digestible. Other sources say that peeling vegetables prior to cooking is a bad thing for retaining fibre, other say that steaming or microwaving is better than boiling. Yet others say that cooking some kinds of vegetables actually helps to make the fibre more accessible to the digestive tract.

I would say this - source say that artichokes and sweet potato are good sources of fibre - and nobody in their right mind eats either of those foods raw - so they must be recommending them on the basis of them being cooked. Hence, cook them for the least time you can - and don't boil them. SteveBaker (talk) 14:01, 24 February 2014 (UTC)

I can't comment on the chemistry or "ruining fibers", but, culinarily speaking, most sources on cooking sunchokes caution against overcooking (even our article mentions it). Depending on the cooking method, some sources say they will turn very tough, while others say they can easily turn to mush. In my experience, sunchokes are not treated like other common root vegetables, for the purposes of cooking: [1] [2] [3]. SemanticMantis (talk) 16:30, 24 February 2014 (UTC)

geography

how many rivers that are said to be in BRAZIL — Preceding unsigned comment added by Andrew Nyabenda Salus (talk • contribs) 11:00, 24 February 2014 (UTC)

http://en.wikipedia.org/wiki/Category:Rivers_of_Brazil 217.158.236.14 (talk) 12:08, 24 February 2014 (UTC)

That category is useful, but likely to be severely incomplete. It includes a few links to "List of rivers" (by subdivision), but even these are probably incomplete. The definition of "river" is somewhat vague - when do you stop calling something a creek and start calling it a river? What about seasonal rivers and underground rivers? If canals don't count, what about heavily canalised but originally natural rivers? Anyways, given the size and climate of Brazil, I doubt that anyone knows the exact number of rivers for any definition of the term based on natural features - though one could always define a river by law as being something on a given list. This might even make sense for the purpose of offering flood protection or regulating pollution or fishing rights, but is likely to leave naturalists unhappy. --Stephan Schulz (talk) 13:07, 24 February 2014 (UTC)

Given the size and general wildness of the place, I'd bet that by far the majority of rivers in Brazil don't even have names. So we're unlikely to be able to find a handy list of them! Wikipedia lists are only allowed to contain items that are considered "notable" - so some tiny tributary someplace is not likely to be listed unless it's famous for something, or widely written about. We can't answer this question that way.

You might expect that the answer for a more heavily inhabited country (like the USA, for example) would be more exact - this article says that the USA has around 250,000 rivers - but our article List of rivers of the Americas says that there are only 34 "major" rivers. So clearly we're not going to be able to do any better for a less-well-documented country like Brazil. If I had to hazard a guess...our List of rivers of the Americas says that there are 34 major rivers in the USA and 12 in Brazil...so about a third the number there are in the USA. If the number of minor rivers is proportional to the number of major ones - then we might hazard a guess that there are about a third of 250,000 rivers - or about 80,000 in Brazil. Another way to do this estimate is by land area. The USA covers about 9.8 million square kilometers - and Brazil covers about 8.5 million...about 87% of the area of the USA. On that basis, we might guess that the number of rivers would be about 87% of 250,000 - maybe 217,000. These two estimates are obviously wildly different - which is a measure of how bad any naive guesses are likely to be!

Short answer, we don't know - and realistically, there is probably no way to know...but if you want a wild guess...between 80,000 and 220,000 is a reasonable range. SteveBaker (talk) 13:45, 24 February 2014 (UTC)

Except that your approxmations aren't likely to hold; given that the U.S. has a very different climate and geography in general than Brazil. Most of Brazil's area is tropical rainforest, which is likely to have many more "minor rivers" (however they are defined) than does the U.S, given that larger proportions of the U.S. are dry plains, deserts, and mountain areas where rivers may be less prevalent. --Jayron32 14:13, 24 February 2014 (UTC)

I completely agree...what I'm trying to say is that we can't really come up with even an approximation. Even knowing that the rainfall in Brazil is higher doesn't help much because that water might end up in a few very large rivers rather than a large number of small ones...I could easily imagine that the more mountainous terrain of parts of the USA would produce more rivers than the boring flat bits. It's really impossible to come up with a good estimate. But if we absolutely have to come up with a number - then we can be almost certain that it's more than a thousand and less than ten million - and we'd probably be safe in saying that it's between 10,000 and 500,000 - and if I were answering a question in a pub quiz, I'd say 200,000. SteveBaker (talk) 14:39, 24 February 2014 (UTC)

Speaking as the person that compiled the "List of rivers" (by subdivision) articles for Brazil they are reasonably complete, even for very remote parts of Brazil, and include anything that had the name Rio X in Portuguese, which would translate to "rivers" in the U.S., they generally do not include smaller streams (ribeiro in Portuguese or creeks in English), but going through and adding up those lists would give you a reasonable approximation for the number of "rivers" in Brazil (that may not be comparable to other total numbers of rivers you find for other countries though, for the U.S. at least those usually do include creeks). Kmusser (talk) 21:31, 24 February 2014 (UTC)

Ah! Well, thanks for your efforts in compiling the list! So the issue here is that the vast majority of the 250,000 "rivers" in the USA wouldn't be considered to be 'rivers' in Brazil (or maybe they would...but not in Portuguese)? I guess that brings us to the cultural matter of how many rivers the local people consider they have versus the number that an outsider (say, from the USA) might consider them to have. Well, that just throws another level of uncertainty into the mix! It's not entirely unsurprising that we have this kind of categorization problem - consider, for example the word "City" - in British English is basically means "a place with a cathedral", so in a population of 60 million people, there are just 69 cities. In the USA, the word "city" means something entirely different - there are 50 cities whose names begin with the letter 'A' in Texas alone and the "city" of Lost Springs Wyoming has only three buildings and a population of 4 people! SteveBaker (talk) 22:04, 24 February 2014 (UTC)

So, as so often with Ref Desk answers, we end up in a morass of terminology and definition issues. What exactly is a "River"?

• According to Wiktionary, it is "A large and often winding stream which drains a land mass, carrying water down from higher areas to a lower point, ending at an ocean or in an inland sea."

• But according to our River article, it is "A natural watercourse,[1] usually freshwater, flowing towards an ocean, a lake, a sea, or another river."

So if reaching the ocean (or an inland sea) is a requirement, then I very much doubt that the USA has 250,000 of them - but if merely reaching another river suffices - then it's easily possible that there are 250,000 tiny streams, creeks and other tributaries that flow into larger rivers before they reach the sea. The word is a complete mess. Webster's dictionary concurs with Wiktionary - but Dictionary.com agrees with our Wikipedia article. World English dictionary says that rivers have to reach the sea...the Collins English dictionary says that it "usually" reaches the sea - but the Oxford English Dictionary says that rivers can flow into other rivers. SteveBaker (talk) 22:04, 24 February 2014 (UTC)

Note that e.g. the Rio Negro is a tributary of the Amazon, so while a river by the argument "rio means river in Portuguese" (and, to be honest, by any reasonable standard ;-), it's apparently not a Webster river. Neither is the Missouri River, or, arguably, the Blue Nile. I think I prefer the somewhat vague Wikipedia definition. --Stephan Schulz (talk) 22:52, 24 February 2014 (UTC)

There is absolutely no agreement on what a river actually is so any effort at describing how many of them there are is essentially futile! SteveBaker (talk) 22:03, 24 February 2014 (UTC)

A simple example is the crazy river system in New York City. The origin of the word "river" highlights its inherent ambiguity.[4] ←Baseball Bugs ^{What's up, Doc?} carrots→ 22:57, 24 February 2014 (UTC)

Can't we just count them from a satellite picture? Presumably, if we only count rivers with a width larger than u, then this number will satisfy some simple scaling law if we let u go to zero. You then need to estimate an effective lower cut off for u (if you take u too small the simple scaling law will fail, obviously you don't have rivers with a width of a micrometer). Count Iblis (talk) 00:06, 25 February 2014 (UTC)

1) I don't share your confidence in the existence of said scaling law, except perhaps at continental scales where topography might average out. (What is this law anyway, functional dependence of number streams of (average) width u per m^2? How do you deal with stream identity for counting purposes, in cases of merging/splitting? ) 2)I don't think such a law would "fail" in the case of tiny rivulets. If done right, we'd expect such a law to put the number of streams to approach infinity as u->0, right? SemanticMantis (talk) 00:48, 25 February 2014 (UTC)

Geologists and hydrologists don't usually bother defining "river" with precision. As discussed above, there are problems with most pragmatic definitions. Politically, navigable waterways are sometimes carefully defined in some jurisdictions. Stream is the all-inclusive scientific word, though it usually does not include situations where water flow is intermittent or seasonal. Count Iblis has the right idea above: pick an average minimum width and length, then count. SemanticMantis (talk) 00:42, 25 February 2014 (UTC)

Nothing to really add to that other than to confirm that there is no generally agreed upon definition of river. The definition I used in making the lists was if it had River (Rio) in the name than it was a river. The USGS avoids the issue entirely by calling all watercourses regardless of size "streams". Kmusser (talk) 17:27, 25 February 2014 (UTC)

• Has no one asked "How Long Is the Coast of Britain?" μηδείς (talk) 17:18, 25 February 2014 (UTC)

Coincidentally, I did wonder: If all 250,000 "rivers" in the USA drained into the sea - then how far apart would their outlets be? I briefly wondered how long the coastline of the USA is...but then sanity returned! Since the coastline of the USA is infinitely long, there is plenty of space for all of them. :-) SteveBaker (talk) 20:58, 25 February 2014 (UTC)

Augmented reality to make a large screen?

I was thinking that Google Glass and competitor products might be able to make a blank wall seem like a really big display. Would that be possible without being able to track eye movements? --129.215.47.59 (talk) 17:59, 24 February 2014 (UTC)

Well, you'd need two things to get it done, in my opinion:

1) 3D technology. This can be done by showing slightly different views in each lens. Of course, the film must have been recorded in 3D. There are some methods to try to convert 2D film into 3D, but they don't work that well. (If it could track your eye movements it could then add in a focus option, to make things sharper when you focus at that depth and direction, and make everything else blurry.)

2) High resolution. How high I'm not sure, but probably at least 1920×1080 (and Google Glass is only 640×360). Otherwise the graininess of the image would ruin the effect. StuRat (talk) 18:05, 24 February 2014 (UTC)

If I understand correctly, I believe the idea is more like having something like a green-screen on a wall, and when viewed with special Ubermajik glasses, becomes a (2D?) display. ~:71.20.250.51 (talk) 19:52, 24 February 2014 (UTC)

This kind of thing used to be done in flight simulators before the advent of relatively cheap 3D graphics hardware - they'd track the motion of the user's head and eyes and project graphics onto the inside of a large dome - but only where the person was looking. This enabled a relatively small amount of graphics hardware to cover a complete 360 degree field of view. There are many "gotchas" though - one is that recalculating the view to display it somewhere else takes time - so the system can't measure where the user's eyes are looking NOW - instead it has to predict where the person will be looking a few tens of milliseconds into the future. Another problem is that our peripheral vision has to be filled - so it was necessary to draw a very low resolution display over the entire area and just inset the high resolution part where they were looking.

Google glass only creates a tiny patch of graphics in the top-right corner of your field of view - and it doesn't track your eye positions - only your head as a whole - so it really can't do this.

You should probably look at something like the Oculus Rift for a cheap, modern "virtual dome" display. Because it's headset more or less fills your entire field of view, it doesn't need to track your eye motion - only your head (which is a much easier thing to do). The Oculus can easily provide a simulation of a display of any size at almost any distance you want. SteveBaker (talk) 20:30, 24 February 2014 (UTC)

I agree that the Rift is the closest thing (or will be, when it comes on the market), but the initial versions won't be all that terrific in terms of resolution. Looie496 (talk) 20:53, 24 February 2014 (UTC)

Our article says the consumer version will have ≥1920×1080 resolution, which might be enough so you don't see the pixels (although if they mean half that res on each eye, that's probably not enough). StuRat (talk) 02:00, 25 February 2014 (UTC)

Yes, that resolution will be split between the two eyes - so each eye gets 860x1080 (actually, rather less than that because some of the screen area is outside of the visual field of the lenses and is therefore wasted - and the distortion of the lenses requires software to map that resolution unevenly across your field of view - so the worst-case resolution is considerably lower than that number). The total horizontal field of view is around 90 degrees (or at least it was in the developer version that I worked with) - with relatively little overlap between the left and right eye. So each eye sees maybe 50 degrees of horizon. With the developer version (which was 1280x800 - giving 640x800 to each eye) you can't really see individual pixels - but the lack of resolution makes everything a little blurry. Comparing that with viewing a TV or computer monitor is a little tricky because it depends on the screen size and the distance you are from it. The monitor I'm currently using is 1920x1080 with about a 50 degree field of view at my normal typing distance - so the consumer version of the Oculus has about half of that resolution. It won't look as sharp as a typical large-screen TV viewed from across the room.

Some of the head-mounted displays that I used a decade ago while working in military flight simulation have vastly better resolution than the Oculus - but they cost about the same as a really nice Italian sports car. The Oculus isn't really the awesome technological breakthrough that a lot of people think it is - what it mainly does is to shrink the cost down to something affordable - albeit with quality that just rides the edge of consumer-grade acceptability.

SteveBaker (talk) 14:16, 25 February 2014 (UTC)

What kind of res did the military stuff have ? If we apply a cost version of Moore's Law to the price of that military grade hardware, it might be down to the range a consumer could afford by now. Or, alternatively, if we apply it to the Oculus Rift, it's resolution might be up to consumer expectations in a few years.

The Oculus also seems to be focused on games (and at least one sex simulator), but I can imagine many more uses, such as virtual tours of houses for sale, eliminating the need to visit any which clearly don't meet your needs. It could also make teleconferencing more interesting, if you had a video camera/screen in your seat at the conference, rotating around as you move you head. StuRat (talk) 19:45, 25 February 2014 (UTC)

It's hard to equate the two. The military headsets that I worked with were designed to be bolted onto a standard flight helmet - so they were huge and heavy. Rather than using a direct-view flat panel display, they used tiny projectors that lit up large curved mirrors in front of the pilot's face. The screens had some kind of exotic coating on them - so if you touched them with an un-gloved finger, you'd permanently mess them up. They also tended to work at higher-than-normal video refresh rates for reasons related to helicopter pilots and strobing effects from the simulated rotors. Generally too specialized and too fragile for the consumer market.

The Oculus approach is pretty much all that works - large flat-screen displays are getting cheap because they go into tablet computers, so it's inevitable that you'll be using one of those...you'd like to have one for each eye - but that doubles the price of the system, so it's unlikely to happen no matter how cheap you can go. The head-tracker isn't anything special - they did have to engineer the interfaces to minimize the latency through the thing - but otherwise it's pretty standard stuff. The real pain is the optics - and those have to be engineered to be both cheap - and to work for people with less than 20-20 vision - and to fit into a space that can reasonably be strapped to someone's face without being too heavy or unbalanced.

Despite all of that care, I'd say that at least 70% of the people at work who were working with them got motion-sick after 10 to 15 minutes of wearing them - and plenty of people suffered headaches and dizziness after prolonged sessions with them.

Head-mounted displays have a way to go yet! SteveBaker (talk) 21:46, 25 February 2014 (UTC)