Wikipedia:Reference desk/Archives/Science/2016 October 31

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

October 31

Balloon descent

How does a large helium balloon (e.g. a blimp or something the size of a hot-air balloon) descend? Hydrogen isn't particularly rare, so you could probably just let off some of it without losing a lot of money, and with a hot-air balloon, you can presumably just stop heating the air, but re-filling the Goodyear Blimp with helium sounds absurdly expensive. <helium balloon descend> on Google is all about party balloons and the like. Nyttend (talk) 03:02, 31 October 2016 (UTC)

See Ballonet. -- ToE 03:18, 31 October 2016 (UTC)

Modern dirigibles are usually flown a bit heavy (more weight than buoyancy). They use engine power to get off the ground and if the engines stop they sink slowly back to the ground. Using downward thrust gets them back to the ground faster. If for whatever reason the dirigible is light (more buoyancy than weight) and the engine isn't powerful enough to get it down to the ground, the only option is venting some expensive gas.

Balloons, which have no engine, can only descent by venting gas. Our article on gas balloons mentions that they are often (in particular in Germany) filled with hydrogen. There are also the Rozière balloon, which is a hybrid, using both lighter-than-air gas to get most of the buoyancy and hot air for altitude control and to compensate for changes in the temperature difference between the lifting gas and the outside air.

In principle, an aerostat can also descent by collecting rainwater or by using a chemical that gets heavier by reacting with air, like hydrocarbon fuel if you collect the water vapour, but that is not very practical for manoeuvring. Finally it's possible to take gas out of the balloon and compress it into a cylinder, but as far as I'm aware of that isn't really used because it takes quite a lot of weight and energy. PiusImpavidus (talk) 10:06, 31 October 2016 (UTC)

• Condensation of exhaust was used on some US airships, such as the USS Akron. Germany also used blau gas, as this could be stored at a similar density to air, thus didn't affect buoyancy much as it was burned.

Zeppelins were much safer when filled with hydrogen. Although the fire risk is obvious, this is over-emphasised after the Hindenburg. Far more accidents were caused instead by loss of altitude control, and this was often related to the difficulty of controlling buoyancy without venting helium. Len Deighton's Airshipwreck is a good study of the hazards.

The last fatal airship accident was another fire, but this time to one of the Goodyear blimps - helium filled. Andy Dingley (talk) 11:20, 31 October 2016 (UTC)

Death following crush injury

In her book, The Good Women of China by journalist Xue Xinran, Chapter 5: "The Mothers who Endured an Earthquake" - referring to survivors of the massively destructive Tangshan earthquake (28 July 1972 1976). It includes a mother's 1992 account (17 years after the event) of her 14-year-old daughter, "the lower half of [whose] body was wedged between the reinforced concrete slabs of the wall...", and later findings that "her legs had been crushed to a pulp...[and] her pelvis had broken under the pressure." For two weeks, "food and medicine were brought to her every day and someone came to nurse her..." and she was "exposed to the elements" while futile efforts to free her continued. The girl attests to "numbness" without pain in her lower body and is able to speak and even sing to the gathered crowd before she expires "after fourteen days and two hours."

My question: is it plausible to survive two weeks trapped with a crush injury?* In the widely publicized case of 13-year-old Omayra Sánchez in Armero, Colombia, trapped by mudslide debris in the aftermath of a 1985 volcanic eruption, she survived a documented 60 hours. Despite there being no way to compare the nature and extent of the two girls' injuries, my query is about triage and treating survivors of crush injuries, while they're trapped and after their extrication, given the possible complications of crush syndrome along with other factors facing rescue teams. *(My intent is to expand the page content.) -- Deborahjay (talk) 10:23, 31 October 2016 (UTC)

Yes. It's also plausible to die very soon after being released from it.

Survival rates increased a lot in the early part of WWII, following experience of treating the large number of civilians injured in the London Blitz. Rhabdomyolysis is a condition that may be treated relatively simply, with the right knowledge, but is otherwise likely to lead to kidney failure, and then death some time afterwards (chronic renal treatment is generally limited in wartime). Andy Dingley (talk) 10:46, 31 October 2016 (UTC)

OP clarifies: Particularly, how long can a person survive with lower extremities pinned by debris, i.e. vital organs unimpaired but field amputation impossible. -- Deborahjay (talk) 11:13, 31 October 2016 (UTC)

: Deborahjay, it was in 1976, not 1972. -- Jack of Oz ^{[pleasantries]} 18:15, 31 October 2016 (UTC)

Thanks; corrected above. -- Deborahjay (talk) 11:08, 1 November 2016 (UTC)

We should then discuss the precise cause of death. Since the lower limbs are not vital organs, their loss is not, in itself, fatal. However, lacking circulation will cause the flesh to die, then turn gangrene, and cause death by sepsis (blood poisoning). Depending on how much blood flow remains to the legs, this process could be very slow, indeed. A pair of tourniquets may help slow the spread of the sepsis from the legs, but this doesn't seem practical if the hips are crushed, too. A transfusion could buy some time, too, if the infected blood is discarded and replaced by fresh blood of the proper blood type.

Of course, death from dehydration would occur much sooner, if no water was available. And exposure to the elements could also cause death sooner, but that would depend on the weather and what measures were taken to protect her from it, such as blankets to protect from cold. StuRat (talk) 04:19, 3 November 2016 (UTC)

Exact Length of Terrestrial Meridian

I am interested in the error margin contained in the WGS84 estimation of a longitudinal quadrant at 10,001,965.729 meters, corresponding to a meridian of 40,007,862.916 meters. I ask this for three main reasons:

• Measuring something as `brutal` as that with a precision of millimeters seems ludicrous.
• Meridians might not all have the same length in the first place. I've read that the concept of a geoid might be related to that, but I'm not sure.
• I've heard a newer estimate of roughly 40,009,152 m (albeit I am unsure whether this might not refer to the length of the Paris meridian: see above).

Can anyone provide me with current or updated values for the measurement uncertainty inherent such length estimations ? — 79.113.212.81 (talk) 12:11, 31 October 2016 (UTC)

Clarification: I am curious about how the meter would have to be redefined in terms of the speed of light so as to make the length of a (mean ?) terrestrial meridian match the original historical definition of 40,000 km as close as humanly possible. This has been the meaning and intent of my question. I've calculated c = 299,733,5381⁄2 ⁠m^*/s⁠ , where m^* represents the `new` or `meridional` meter. However, such precision would be absurd in light of the above. I am curious about the range which the value of c can span (not in the sense that c actually varies, but in the sense that the same constant can be numerically expressed in various ways, depending on the actual range within which our knowledge of the terrestrial meridian can vary). — 82.137.53.158 (talk) 20:15, 31 October 2016 (UTC)

Further Clarification: I honestly couldn't care less about either WGS84 or global positioning systems. (The only reason I even mentioned its name was because I gave a number, and it seemed common sense to mention the source of one's information, that's all. Apparently, this was a complete mistake, since it completely detoured the focus of the conversation). The only thing I care about is determining the smallest possible interval on which the length of terrestrial meridian(s) is known to exist with absolute certainty; e.g., something like 40,007.5-40,009.5 km, but I'm not sure if this is either the smallest, OR if it's limits are even 100% certain. — 79.115.135.102 (talk) 11:20, 1 November 2016 (UTC)

WGS 84 measures the reference ellipsoid of the geoid, not the Earth itself. It's not so unreasonable to measure this to mm precision. Andy Dingley (talk) 12:30, 31 October 2016 (UTC)

Not unreasonable, but certainly useless, not to mention confusing. — 79.113.236.215 (talk) 13:20, 31 October 2016 (UTC)

GIS specialists spend years studying applied mathematics and advanced software - the reference geoid is completely reasonable and useful. It just requires some domain-specific knowledge. Did you know that you can get a master's degree in geographic information systems from multiple highly-competitive, prestigious universities? ... or that geographers earn well above the median salary in the United States, even at the start of their careers? So ... when you encounter a topic that is a bit complex, don't write it off as useless... confusing, perhaps, but our marketplace economy seems to think that it's very useful. The reference geoid may seem like a bit of "technobabble," but it's a crucial part of engineered systems you use daily, like internet mapping services and global positioning system electronics and software. It just requires some advanced mathematics. Here's a nice website from NOAA on the topic: Converting GPS Height into ... Elevation ... Using the Geoid. Nimur (talk) 14:26, 31 October 2016 (UTC)

That... was not what I was talking about. All I meant was that... (how to explain this ?) the value in question is the direct result of applying some formula to certain inputs... but the inputs themselves have a certain degree of error... and it would be absurd to pretend to evaluate the output with a higher relative accuracy than the imperfection inherent in the measured inputs themselves. — 82.137.53.158 (talk) 15:54, 31 October 2016 (UTC)

If you can't make the measured inputs any more precise (and the earth moves over time, or with tides, so you can't) then don't use them. Invent an arbitrary reference instead, which is what the ellipsoid does. From this you can then make higher precision measurements. Most of the reasons to make such measurements aren't about geodesy, they're about arranging two more arbitrary pieces of equipment - you don't care much about either of their relationships to the Earth, it's about their own relative positioning. Take long baseline astronomy as an example. Andy Dingley (talk) 16:22, 31 October 2016 (UTC)

I think the OP is concerned about the digits after the decimal point. I was taught that we should only report means and standard errors/deviations to the accuracy of the instrument used to measure the variable. So, if I used a ruler with cm and mm divisions to measure tail length of a group of lizards, the mean might be 3.4 ± 0.4 cm. If the OP is looking for a convenient and accepted method of reporting large numbers, they should take a look at Scientific notation. DrChrissy ^(talk) 16:42, 31 October 2016 (UTC)

The issue with the idealized meridian used in WGS-84 is that there is no device used to measure it. It's based on an idealized geometric shape, not on the actual earth, so hypothetically, it has infinite precision. --Jayron32 17:44, 31 October 2016 (UTC)

The WGS 84 reference ellipsoid is a defined coordinate system. Hence, the characteristics of that ellipsoid are exact (infinite precision), by definition. However, the reference ellipsoid is useful precisely because it is intended to be a reasonable approximation to the surface of the Earth. One can ask, how similar is the length of a meridian on the actual Earth to the length of the meridian on the reference ellipsoid? That's actually a trickier question than it might appear. How does one define the length of a real world meridian? Which meridian? Do you include changes in elevation at the surface, or do you try to measure an equivalent distance if the path was fixed at sea level? If you want to travel over the surface of the real ocean / land, then your result depends not only on which meridian you study but also on what time you measure. Tides (both water and solid earth) will affect the result by a small amount, as will small alterations to the landscape that occur over time. It would be tricky to specify a meridian distance based on the real world that is well-defined, let alone measurable, at the millimeter level of precision. More practically, if one chooses to follow the actual geometry of the Earth, all those little ups and downs are likely to add many meters to the distance traveled. That said, is the meridian distance actually useful? The WGS 84 coordinate system is definitely useful, and widely used. By contrast, measuring a physical meridian would seem to have few, if any, practical applications. Dragons flight (talk) 18:25, 31 October 2016 (UTC)

By the way, Earth_ellipsoid#Historical_Earth_ellipsoids provides some examples of historical references ellipsoids and how they have changed over time. This might be used to infer some sense of the uncertainty in the defining parameters. As before, any specific coordinate system is by itself simply a defined structure, but we have changed the definitions over time to try and better approximate certain features of the real world. Exactly which features one most tries to capture have also changed over time. Dragons flight (talk) 18:33, 31 October 2016 (UTC)

The reference ellipsoid is indeed intended to be "a reasonable approximation" to the Earth. That's the point at which the approximation is made, not in the definition of that ellipsoid. There are purposes where positioning more accurate than the Earth's shape is valuable, and so is achievable. Andy Dingley (talk) 19:05, 31 October 2016 (UTC)

Global Positioning System: Theory and Practice [1] gives four values with 7 or 8 significant digits, including the semimajor axis a = 6378137m. This is the offically used estimate and given as 6378137.000000m in contexts with decimals. The book then says: "... the semiminor axis (b = 6356752.314 m) can be derived." The value is rounded from a calculation and it varies how many decimals (if any) are given. World Geodetic System#A new World Geodetic System: WGS 84 says 6356752.314245m in WGS 84 and 6356752.314140m in GRS 80, so nearest mm isn't even enough to distinguish them. Nobody claims the actual Earth is within 10 mm of 6356752.314m, but if the value is rounded to an integer meter like the estimate for a then it probably causes problems in some applications where it isn't sufficiently close to the mathematically expected relationship to other values. Consider for example two close objects which must match to a millimeter, but one uses rounded b = 6356752.000m while the other uses the four values used to derive b, corresponding to assuming b = 6356752.314140m. 10001965.729m is a rounded value derived from a = 6378137.000000m and b = 6356752.314140m. PrimeHunter (talk) 02:26, 1 November 2016 (UTC)

In 1791 a quarter of the meridian i.e. the distance from the Equator to the North Pole via Paris, was declared 10 000 000 meters exactly by definition. Since then the actual physical distance is not known to have changed and it has been studied since 1795 by the French Bureau des Longitudes. They do not claim that measurements on the ground with millimeter repeatability are possible. What has happened is that the meter standard was changed by declaration in 1983 to the distance travelled by light in vacuum in 1/299792458 of a second, equivalent to saying light travels at 299792458 meters per second exactly by definition. It is the deliberate exactness of these integers that allow an appearance of unrealistic precision when one makes an estimate in "new" meters of the meridian by Least squares regression of the best available data. AllBestFaith (talk) 00:53, 1 November 2016 (UTC)

I wasn't claiming that the purpose of the 1983 redefinition was to determine the meridian length to millimeter precision(!). Nevertheless, I want to know on what interval the length of a meridian varies. 40,000 km is definitely too small, and 40,020 is definitely too big; but can we say with absolute certainty that the meridian is strictly bigger than 40,005 km ? How about 40,007 km ? Or even 40,007.5 km ? Likewise, can we say with absolute certainty that the meridian is strictly smaller than 40,015 or 40,010 km ? How about 40,009.5 km ? (This is what I had in mind). — 79.115.135.102 (talk) 11:20, 1 November 2016 (UTC)

Which meridian, and how are you defining it? As I said above, the length of a real world meridian is going to depend on exactly what you are measuring, and potentially also on when you measure it. The distance on an idealized ellipsoid is of course different from any distance measured over the physical Earth. Dragons flight (talk) 11:42, 1 November 2016 (UTC)

Variations are indeed imminent. But I do not want to eliminate them, I want to find out what interval they span. Can we, for instance, say with absolute certainty that all meridians (or at least the ones from London and Paris) are bigger than 40,000 km (or some other more precise value) when measured at sea level, even when the latter is at its lowest ? Or, conversely, that they are smaller than 40,010 km (or some other more precise value) when measured at sea level, even when the latter is at its highest ? — 86.125.203.53 (talk) 12:08, 1 November 2016 (UTC)

It is worth noting that there were in fact four definitions for the meter between the meridian definition and the speed of light definition. Three of these were based on constructed physical objects. See Metre#Timeline. Dragons flight (talk) 11:49, 1 November 2016 (UTC)

Yes, I am quite aware of that. And I thought about each of them and their possible range (e.g., the `pendulum meter` varies between 991⁄3 ± 1⁄2 cm, for instance. Now, with your help, I want to find what range the `meridional meter` spans). — 86.125.203.53 (talk) 12:08, 1 November 2016 (UTC)

How smooth is a billiard ball?

Does anyone more familiar than I with cue sports such as billiards, snooker or pool happen to know what the relevant standards are for the sphericity and surface smoothness of a ball? If there's a difference between "match grade" and "typical", then either would really do. I can find standards for diameter, and variations in permitted diameter, but that's not the same thing.

(Then does anyone know how this compares to the Earth? - I'm assuming Everest at 8,848m, Marianas Trench at 11,000m and a prolateoblate spheroid of about 1/300 flattening) Andy Dingley (talk) 18:36, 31 October 2016 (UTC)

• You mean oblate spheroid, of course. --76.71.5.45 (talk) 20:38, 31 October 2016 (UTC)

Thanks, I've been listening to that pear-shaped De Grasse Tyson bloke again. Andy Dingley (talk) 20:44, 31 October 2016 (UTC)

You've actually asked a question which has been a popular one, for which there is a lot of references on the internet. The general consensus, FWIW, is that the Earth is smoother than a billiard ball. See This page at Discover Magazine, This set of calculations here, Neil deGrasse Tyson's official twitter account, this thread at SDMB. Everyone once in a while, someone brings up the fact that the earth is not a true sphere, but as NdGT points out, the deviation of the earth from a perfect sphere is also within the tolerances allowed for a regulation billiard ball. --Jayron32 19:00, 31 October 2016 (UTC)

Note however that these references reach conflicting conclusions! The specifications for a billiard ball are in paragraph 16 here and the permitted diameter is given as "2 ¼ (+.005) inches [5.715 cm (+ .127 mm)]". Note that it says +, not ±; I believe this notation means that the diameter must be at least 2.250 (not 2.245) inches, and possibly up to to 2.255 inches. Therefore tolerance is 1 part in 450, meaning that the Earth's 1/300 flattening does not qualify, as Phil Plait says at Jay's first link. The page at the second link quotes the specification correctly, but then interprets it as if it said "±" (which would make the tolerance 1/250), and concludes that the Earth does qualify. The third link states a conclusion without justification. At the fourth link the specification is misquoted as actually saying "±", but one poster in the thread points out that it's describing the permitted diameter of the ball (as if assuming that it's spherical), not how far off spherical it can be. So it's unclear in that respect, but 1/300 variation is clearly too much. --76.71.5.45 (talk) 20:38, 31 October 2016 (UTC)

There are any number of solutions that I've seen to this. However all that I've seen so far are incorrect, as they have confused the diameter range stated in the spec (for regular spheres), with surface roughness, which is the measure that ought to be compared to Earth's mountain ranges. Obviously the range of diameter variation is larger than the acceptable roughness (a 5 thou local variation, i.e. roughness, would feel like fine sandpaper), which means that the "Earth is smoother than a billiard ball" claim is inadequately proven by this comparison. Andy Dingley (talk) 20:42, 31 October 2016 (UTC)

There is some commentary here [2] about the roughness suggesting the earth is definitely rougher than a billiard ball. It isn't an RS but based on someone's personal estimation from balls they have studied although also looks at photos of balls ([3]) and other things. But given that the WPA didn't respond it may be the best you'll find. (Although it's possible the person just got unlucky. And I suspect someone like NdGT is more likely to get a response.) There is also some commentary on the images and other things here [4] which comes to the same conclusion. Also mentioned in the first link is [5] as there are some comments there which may be helpful. (Although there are so many comments there including of other stuff it's hard to find the good ones, but I did see some discussion of smoothness etc. Maybe Paul Hutch's comments are the key ones.)

IMO I would be willing to conclude from these that the earth is not as smooth as a normal/typical undamaged billiard ball. (Similar conclusions are reached about the roundness although I'm a bit confused by this since as mentioned by others, some sources conclude the earth isn't even within the spec for roundness.)

Whether the earth is as smooth (or round) as a billiard ball technically can be I'm not sure. Not legal advice but if the WPA doesn't actually specify smoothness, it may be such a ball is indeed technically allowed. Unless there's some loose policy like the referee being able to decide if a ball is acceptable condition without any definite specification of what this entails. (It'll probably help if both players agree they want a different ball/s which I suspect will normally be the case.) In any event, if people regularly start producing such balls, it may be they'll fix their specification.

Nil Einne (talk) 23:46, 31 October 2016 (UTC)

Maybe I should just find a billiard ball from somewhere and stick it under the profilometer? I'm rather tempted now to print an accurate Earth globe, with relief to scale (I've never seen one at scale), just to see what it does feel like for texture. Andy Dingley (talk) 00:35, 1 November 2016 (UTC)

For librettist W. S. Gilbert the 2-D Ellipse was probably easier to visualize than its 3-D analogue the Ellipsoid, hence this harsh decree:

 The billiard sharp who any one catches,
 His doom's extremely hard —
 He's made to dwell —
 In a dungeon cell
 On a spot that's always barred.
 And there he plays extravagant matches
 In fitless finger-stalls
 On a cloth untrue
 With a twisted cue
 And elliptical billiard balls

- The Mikado, 1885. AllBestFaith (talk) 01:10, 1 November 2016 (UTC)

Can sexual preferences change or be changed?

I don't mean in the sense of a mad spiritual cure for homosexuality, but more in the sense of an exposure to different stimuli. For example, if a man has access to one type of porn or the other, would he develope different tastes? Is there anything like acquired tastes in the sexual field? --Llaanngg (talk) 23:42, 31 October 2016 (UTC)

Well, they do say that cunnilingus is an aquired taste. I coudnt possibly comment.--213.205.253.204 (talk) 00:11, 1 November 2016 (UTC)

Q? What comes after 68? A: 69. Q? What comes after 69? A: Mouthwash. --Jayron32 00:59, 1 November 2016 (UTC)

Sexual preference as a mechanism in evolution, including extreme manifestations in animals of the Fisherian runaway such as male peacock plumage, can only operate if mates are preferred from the accessible population. By this Darwinian argument, tastes are shaped by the existence of potential breeding partners ergo a different taste would be acquired among a different population. However, opposing this are critics of Darwinism who suggest that sex in humans has more a social or spiritual than evolutionary function, and Pornography serves interest in Sexual arousal usually without any motivation to conceive progeny. AllBestFaith (talk) 01:51, 1 November 2016 (UTC)

The "exposure to different stimuli" hypothesis was tested by Sean Thomas in his 2007 memoir Millions of Women are Waiting to Meet You, reviewed here. He would seem to agree with your tentative hypothesis. Carbon Caryatid (talk) 12:53, 1 November 2016 (UTC)

And I forgot to mention Situational sexual behavior. Human sexuality is infinitely variable. Carbon Caryatid (talk) 12:56, 1 November 2016 (UTC)

My feeling is that human pheromones have this effect, but most of the research tends to be on their near-term effect as 'releasers' (aphrodisiacs, essentially), rather than proving a role on long-term changes in sexual preferences and orientation ('primers'). There are, however, occasionally studies pointing out the correlation of sexual orientation with the reaction to pheromones. You may find out more than I know t [6]; see also [7]. Wnt (talk) 09:41, 2 November 2016 (UTC)

Much of human behavior is learned in childhood, and can't be change much once we are adults. In the case of sexuality, that may develop a bit later, perhaps into early adolescence. StuRat (talk) 16:20, 3 November 2016 (UTC)

If you've never heard of something, literally never encountered an idea (not just the word for it, but the concept), then it isn't part of your mental - and by extension sexual - world. Once you have, it is, or can be. One effect of the porn industry is bringing what this Vice article calls "extreme anal practices" to a much wider audience.

What is changing, however, is the encroachment of hardcore sex acts from the fringe into the mainstream. “Everyone’s pressured to do anal,” Sheena [Shaw] says. “Culture teaches us what to like and what not to like.” What used to be taboo is now a staple, and the public is taking notice. On a thread titled “Increased Visibility of Anal Prolapse: Reasons?” from September 2013 on the online forum Adult DVD Talk, a user <snip> hypothesized that anal performers just hid their prolapses off camera in the past. “(They) are now only showing it on screen because it has become acceptable and there is a growing market for it.”

Considering that brain researchers are re-defining "adolescence" as well into the 20s [8], with guidance issued to child psychologists to that effect [9], then it seems entirely reasonable to suppose that sexual tastes mature, or at any rate develop, along with all the other ones - music, fashion, food. Not all adults like gin & bitters, or olives, or Beethoven, but plenty do; very few children have acquired a taste for any of them. Carbon Caryatid (talk) 20:30, 3 November 2016 (UTC)

I think this maturation continues long after even an extended adolescence. I mean, who in their 20s thinks that someone with gray hair can be pretty? Wnt (talk) 11:20, 4 November 2016 (UTC)

Quite a few, apparently: [10]. StuRat (talk) 00:58, 5 November 2016 (UTC)