Wikipedia:Reference desk/Archives/Science/2017 August 20

= August 20 =

Umbra speed on Earth's surface


During a solar eclipse, why does the speed of the moon's shadow on the surface of the Earth vary significantly? —2606:A000:4C0C:E200:A90E:D475:2878:2440 (talk) 01:28, 20 August 2017 (UTC)
 * 2410 mph: Western Oregon
 * 1747 mph: central Nebraska
 * 1462 mph: Western Kentucky
 * 1502 mph: near Charleston SC
 * What is your source for those numbers? ←Baseball Bugs What's up, Doc? carrots→ 02:49, 20 August 2017 (UTC)


 * I assume the time of day should be a factor. Think of a shadow cast on a globe -- if the Sun is just rising, a small movement of a shadow across it will go a long distance on the ground.  Likewise the latitude should have an effect.  Also, the rate at which the ground moves due to the rotation of the Earth will be different by latitude.  The Moon's orbital speed should not vary by much, so the shadow itself (ignoring the terrain and curvature of the Earth in its way) should not change speed much. Wnt (talk) 02:51, 20 August 2017 (UTC)


 * Source:
 * 2606:A000:4C0C:E200:A90E:D475:2878:2440 (talk) 02:55, 20 August 2017 (UTC)


 * I don't think time of day should matter. Essentially, the shadow itself doesn't move (relatively speaking) -- the Earth rotates within the umbra; the time as determined by the position of the sun in the sky (not "clock time") is the same for all observers (for same eclipse phase). 2606:A000:4C0C:E200:A90E:D475:2878:2440 (talk) 03:04, 20 August 2017 (UTC)
 * To excerpt from that source "Because of the geometry of the Earth’s shape, the shadow will travel faster across its surface and the ends of the eclipse path, and slowest right in the middle." (It should say "at the ends". Must be a typo.) Bus stop (talk) 03:07, 20 August 2017 (UTC)
 * I guess the relative speed of the moon (~one orbit per month) does matter. Looking at the top image from that source, the moon's shadow always remains in the "center" of the Earth's surface, since the Sun's direction (vector) must be normal to the surface, with the moon intersecting the vector. 2606:A000:4C0C:E200:A90E:D475:2878:2440 (talk) 03:29, 20 August 2017 (UTC)
 * I can't see the image you refer to, but (speaking as an ex-astronomer) I think you must have misinterpreted it, if not the entire physical situation. During a solar eclipse, the angle to the Earth's surface of the direction to the eclipsed sun can vary from 0° (if the Sun & Moon happen to be at the at the Zenith, which would actually be a rare occurance) to 90° (if they're on the horizon, which happens not infrequently). {The poster formerly known as 87.81.230.195} 94.12.90.255 (talk) 05:52, 20 August 2017 (UTC)
 * I can see now that the alignment need not be a normal vector to the Earth; but, I reiterate my argument that the reason the shadow moves is due to the rotation of the Earth, and the extraordinary change in the velocity of the shadow is curious. -OP:2606:A000:4C0C:E200:5560:5BBD:5A94:28E2 (talk) 08:00, 20 August 2017 (UTC)
 * Umbras start at sunrise and end at sunset. If it goes over the pole before hitting the Earth it can go the "wrong way" (sunset to sunrise). The shadow cannot only move because of the rotation of the Earth, if that were the case there'd be no way to get off the Earth and there'd be a total solar eclipse somewhere on Earth 24/7, even at Full Moon. Sagittarian Milky Way (talk) 09:13, 20 August 2017 (UTC)
 * There always is an umbra; however, it only reaches the Earth during an eclipse. 2606:A000:4C0C:E200:EC30:98E9:F083:2A5E (talk) 17:52, 20 August 2017 (UTC)
 * The rotation of the Earth contributes to the movement of the umbra over the Earth's surface, but is by no means the dominant contributor. Also important are the movement of the Earth due to its revolution round the Sun, the movement of the Moon due to its revolution around the Earth – both of which alter the relative alignment of the syzygy (astronomy) – and the sphericality of the Earth which means the umbra moves faster towards the Earth's limbs (from the Moon's point of view) limbs due to the umbra's increasingly oblique angle to the Earth's surface. Consider setting up a rough 3D model with balls and a flashlight so that you can demonstrate these factors to yourself, rather than trying to visualise them from descriptions and 2D diagrams. {The poster formerly known as 87.81.230.195} 90.204.183.114 (talk) 19:50, 20 August 2017 (UTC)
 * Viewed from an observer at Polaris at this time of year, the Earth is a blue dot (formerly white...) and according to Orbit of the Moon the Moon goes around it counterclockwise. The Earth is going around the Sun counterclockwise, or if you want to be parochial about it, the Sun goes around the Earth counterclockwise also, in a non-inertial frame of course; but either way this is so slow we can almost ignore it.  And since the Moon is able to extract angular momentum from the spinning Earth, it's a good guess the Earth is also spinning counterclockwise, if your scope can make out those details.  So the chief competitors in this race are the Moon, moving an average 1.022 km/s = 2290 mph, and the surface of the Earth, moving 24000 mi/24 = 1000 mph at the equator, times the cosine of your latitude.  (The North Pole at 90 N, of course, does not move when rotating).  So the Moon's speed is definitely faster than the land's speed, and they're both going counterclockwise, so it's going to catch up.  So the East Coast, racing toward sunset, will not outrun the approaching eclipse... unless, that is, it actually reaches sunset first.  But when the Moon's shadow is nearing that side of the Earth, then it is extending further and further back along the side at a faster and faster rate, so it is entitled to a last burst of speed near the end of the eclipse (as near its beginning) to cover a large amount of twilit territory. Wnt (talk) 00:51, 21 August 2017 (UTC)

Somewhat related question
Is it purely coincidental that, as seen from Earth, the diameter of the Sun and Moon are the same? — 2606:A000:4C0C:E200:5560:5BBD:5A94:28E2 (talk) 06:09, 20 August 2017 (UTC)
 * Yep, funny, huh? However, as the Moon is moving away from the Earth, long in the future, this will no longer be the case, and total solar eclipses will no longer occur. --47.138.161.183 (talk) 06:49, 20 August 2017 (UTC)
 * Actually, we already see that, when the moon is closer to apogee and we see an annular eclipse. If the moon drifts farther away, all eclipses will become of the annular variety. ←Baseball Bugs What's up, Doc? carrots→ 00:52, 21 August 2017 (UTC)
 * Hmmmm.... The Moon is thought to have moved from 5 Earth radii away to the present 60. The orbital speed is proportional to the square root of this radius.  So I'm thinking when it was something like 60/(2.3)^2 = 11 Earth radii away, the orbital speed should have been slow enough that the equator could match its speed.  And of course the umbra back then would have been much larger.  Does that mean there was a time early on when eclipses could last for hours, or you could even have a double eclipse where the Sun could peek past the Moon, then return behind it?  Dang that doesn't seem to make sense - the Earth's surface isn't in orbit - yet I rechecked, Earth's rotation is 1040 km/h, orbit of the Moon is 2290 km/h.  Is the orbital speed formula approximation that wrong?? Wnt (talk) 01:54, 21 August 2017 (UTC)
 * The Precambrian day was less than 24 hours. Did you about for that? What did you mean by the last 2 sentences? Sagittarian Milky Way (talk) 02:02, 21 August 2017 (UTC)
 * Wnt, I think you have confused tangential and angular speed. 78.0.199.76 (talk) 03:11, 21 August 2017 (UTC)
 * No, I was just being stupid. The orbital speed is proportional to the inverse square root of radius, i.e. something a billion light-years out has an orbital speed of basically zero, while closer orbits are faster.  So the Moon doesn't slow down to current rotational speed at the equator until it is 6 times further out, if it ever gets there, at which point the rotational speed has also decreased, and there are no total eclipses anyway.  This also reflects that the rotational speed would have to be very much higher than lunar orbital speed to reach orbital velocity near the surface.  Now, I still don't actually know that the faster rotation of the early Earth was never faster than the faster orbital speed of the early Moon, and I can't rule it out logically; I'd need formulas for how both have chanced over time. Wnt (talk) 03:33, 21 August 2017 (UTC)

Runway construction on estuary examples
Are there any similar examples to the Haneda Runway D and bridge project ( https://www.jstage.jst.go.jp/article/jgssp/2/2/2_ESD-KL-4/_pdf) which involves the construction of both a runway and bridge strong enough to carry A380s on an estuary? 90.194.48.37 (talk) 10:58, 20 August 2017 (UTC)
 * As detailed in Thames Estuary Airport, proposals to build a large new airport serving London somewhere in the Thames estuary have been around for over half a century, but so far none have been implemented. {The poster formerly known as 87.81.230.195} 90.204.183.114 (talk) 19:37, 20 August 2017 (UTC)

LED bulbs and resistors
Do LED bulbs need a resistor to work properly? — Preceding unsigned comment added by 123.201.133.189 (talk) 14:29, 20 August 2017 (UTC)
 * Yes, they're in there to limit the current. Unless you use a very weak source like a penlight battery which has enough internal resistance anyway. Some batteries can have very low resistance so unlike the article I wouldn't leave out the resistance even with 3V coin cells. See LED where it talks a little bit about this in the section on considerations for use. Dmcq (talk) 15:18, 20 August 2017 (UTC)
 * LED bulbs need something that limits the current, but that something does not have to be a resistor. Unless the product is really cheap, it is far more common to use a tiny Switched-mode power supply because the batteries last a lot longer. --Guy Macon (talk) 17:33, 20 August 2017 (UTC)
 * There are two aspects to this: controlled current or controlled voltage. Unlike an incandescent lamp, LEDs have a strongly non-linear relation between current and their junction voltage. So a simple resistor isn't a great solution. It's important to control (or at least, safely limit) both aspects.
 * For decades, LEDs were low-powered dim indicators, not powerful illuminators. So their power consumption was low and they were fed from a broadly constant voltage supply. So the simple series resistor was adequate. Also the LED forward voltage (typically 1.7V for a red) was low (compared to the supply), so the resistor was essential. This had both advantages and a disadvantage, also it's somewhat unclear whether the resistor is there to control voltage or current. Because the supply voltage was so much higher than the LED forward voltage, there was a large voltage dropped across the resistor making the circuit relatively insensitive, i.e. the LED current changes little if the supply voltage changed (a good thing). The other advantage of the resistor is that it's cheap and simple, the disadvantage is that the power is mostly dissipated uselessly in the resistor, not the LED.
 * It's necessary to control both current and voltage - or at least, to not exceed limits on either, by controlling one of them. For these early LEDs, controlling the current was enough, and was easy.
 * Then the 1990s changed everything - LEDs became powerful sources of illumination. Also their forward voltages increased, to maybe 3.6V. We were now trying to make powerful LED lighting, also using LEDs as torches, from low battery voltages (maybe just a single cell at 1.2V) and trying to cope with battery voltages that changed as the cells discharged.
 * The series resistor was no longer an adequate control mechanism. As the supply voltage approaches the LED voltage, the circuit no longer regulates so well. But the supply voltage was also reducing at this time, both because the junction voltage was increasing and approaching battery voltages, and because powerful LEDs (especially with batteries) could no longer afford to waste power across the ballast resistor, as they had before.
 * Constant current supplies are now popular for lighting, because they are easy to control and they're also usable with varying numbers of LED chips in a series circuit. However they need a supply voltage that's close to the LED junction voltages (or the difference ends up as waste somewhere), so they might require an embedded switch-mode buck converter (a downward voltage converter) to efficiently match these.
 * Constant voltage supplies are also used, mostly when driven by batteries. These are boost converters, which can raise the voltage, thus allowing batteries to be used as their voltage falls under discharge. However voltage regulation is difficult, as the LEDs are sensitive to their voltage, and a small excess voltage can cause such a great over-current that the LED overheats and is destroyed. Andy Dingley (talk) 18:39, 20 August 2017 (UTC)

-- Hans Haase (有问题吗) 10:27, 22 August 2017 (UTC)

Saving wild animals like in Operation_Breakthrough
This question leads to a broader question: is it a good idea to save animals from natural threats? Shouldn't the evolution just do its work? That is, I wonder why save the whales at all in this situation, shouldn't we better let thoughtless stranded whales die?--Hofhof (talk) 16:07, 20 August 2017 (UTC)


 * "Is it a good idea...?" and "Shouldn't" both implicitly rest upon some value system. Maybe you don't want to "interfere", but surely other people do think there is good to be done by saving a beached whale. Could be something about how humans are pretty much 100% responsible for endangering wales, and maybe some people think we owe them a little help when we can. Nobody can tell you what you think is right or wrong, but we can give you references that discuss what experts in biology and ethics have said about it.
 * For whales specifically, see this report on whale welfare and ethics, this  position piece on beached cetaceans from the WDC, including decision making.
 * For general background, See Whale_conservation, Marine_mammals_and_sonar, Value theory, bioethics, Conservation_(ethic). SemanticMantis (talk) 16:37, 20 August 2017 (UTC)


 * From it appears that this might be a matter of mission creep -- NOAA apparently more often rescues whales tangled in manmade objects, and I would assume the implication there is that they don't expect the sea to be full of crap for the next 10 to 20 whale generations to make it a useful thing to select against.  Whether that's because the crap gets cleaned up or the whales die out before then is yours to decide...  Of course, the literal answer to your question is that it's a great idea: a million dollars in suckers' money waiting to be spent, and in the process you make yourself a star "real life hero" in two weeks of paid work. Wnt (talk) 00:36, 21 August 2017 (UTC)


 * Another possible argument for saving a stranded animal is if it's species is endangered. Of course, it could also be argued that it has defective genes, and should be eliminated for the good of the species.  But this must be weighed against the reduction in genetic diversity that will occur if that individual dies. StuRat (talk) 00:54, 21 August 2017 (UTC)


 * Laymen typically highly overestimate the effects of selection pressure in driving evolution. Typically, most deaths are entirely accidental in regards to any specific gene.  Rather, even if only .01% of deaths are differentially caused by a gene per generation (that is, a surplus of only 1/10,000 deaths) it is enough to drive evolution in geologic time, even though very few excess deaths can be attributed to that gene.
 * Consider the extinction of the non-avian dinosaurs. There was no specific gene that the Chicxulub Meteor selected for that day.  Likewise, very few people who die who have blue eyes die because they have blue eyes.  Also, people who die in old age have already reproduced.  Selection pressure doesn't apply at that point, hence the plethora of diseases of old age.
 * Finally, there's the example Stephen Jay Gould gave of a single dog which killed over 500 kiwis in 1988, about half the local population. There was no single gene these birds shared that the massacre survivors didn't.
 * This is Gould's hypothesis that much of evolution is driven by radical contingency (i.e., blind luck), not gene selection. See Gould's Wonderful Life.  If most whale strandings are due to blind misfortune, attempting to save the whales is unlikely to contribute to their long-term extinction.
 * There are, of course, counter-examples. Should world civilization fall, the bloodlines of type I diabetics, hemophiliacs, and women whose hip structure allows only Caesarian births, for example, would quickly plummet to almost negligent numbers. That is because we address the effects of these genetic issues without eliminating the genes themselves from our own population. μηδείς (talk) 16:24, 21 August 2017 (UTC)
 * A small nitpick. In the Māori language, the plural of kiwi is kiwi. One kiwi, 200 kiwi. One Māori (person), 20,000 Māori. One waka, 30 waka. By convention, this is carried over into New Zealand English when using Māori words. There are many New Zealanders who uncaringly add the 's', but the preference is to follow the Māori style. Akld guy (talk) 20:29, 21 August 2017 (UTC)


 * I bet the Chicxulub Meteor selected pretty heavily for some hibernation/estivation variants, and blue eyes could be the difference between life and death under the Nazis. But in order to be effective, selection has to be sustained - the meteor of course could wipe out whole species, but if it left a few copies of a given allele within a species, that allele might eventually return to its original equilibrium.  Despite all these things, evidence of selective sweeps is not hard to find.  Even if a given gene only accounts for 1/10000 deaths ... there are 30,000 of them.  One gene usually matters little, but some gene usually matters. Wnt (talk) 19:52, 21 August 2017 (UTC)


 * The birds are normally called kiwis (pl) in American English, this is the term Gould uses repeatedly in Bully for Brontosaurus recounting the story I referred to, and I am sticking to it. You can call moose "elk"; rabbits "coneys"; and opossums "possums" as much as you like.
 * As for selection for estivation or hibernation as a means of surviving K/T level events, either life at the time would have had to be small due past such events, or have become small afterwards to survive regular future such events. Neither was the case, so this was one mass culling, not the evolution of small size by (repeated) natural selection.  Dinosaurs did not evolve to become extinct and mammals did not evolve to stay small to avoid future mass extinctions.
 * Each of you did actually understand my argument and nitpicking for points is not my cup of cricket. μηδείς (talk) 00:39, 22 August 2017 (UTC)
 * Don't see any evidence, Akld guy was "nitpicking for points", not sure Wnt was doing so either. Most logical conclusion would be that the person who assumes nitpicks are for points probably believes so because they aren't being quite honest in what their cup of cricket is hence adding non existent motivations just because it's what they normally do. Incidentally, in modern contexts it's not very common to call opossums as possums in NZ. Mostly of course we don't talk about opossums much since they don't really concern us, but possums do. Our article Opossum suggests it's an American thing. There is some history of the possum being called opossum in NZ, including in legislation  but from my experience in modern contexts very few Kiwis actually do that. (I don't think I've actually read or heard any example in the years I've been here except for those I read when researching this. I.E. whenever someone has mentioned opossum they're been talking about opossums although as said, this isn't really that common in NZ. But I have read and heard about possums. A lot. And although it isn't something I have that much experience with, a Google Scholar search supports my view that this applies even to scientific papers in modern contexts. Most references to opossum relating to New Zealand since 2000 seem to be titles of older citations. The modern papers themselves use possum. This source suggests it was probably about in the late 1970s and the 1980s [//books.google.co.nz/books?id=YDM0hjAwchUC&lpg=PT16&ots=4fScLzyt4L&pg=PT16] that the scientific usage caught up with the vernacular usage.) Nil Einne (talk) 08:52, 22 August 2017 (UTC)
 * Basically TLDR, Nil, but both those editors understood my point exactly, and I said they could call opossums "possums" if they liked. All that matters is they were making points irrelevant to the actual science.  Mass extinction does not equal evolution, although it obviously opens up niches that other animals can evolve to fill--but mass extinction and evolution by natural selection are separate concepts.  As for PC nomenclature, they are still starfish and jellyfish no matter how much the mermen complain. μηδείς (talk) 16:36, 22 August 2017 (UTC)
 * I have no idea what relevance moose, elk, rabbits, coneys, and possums had to my little piece about the plural of kiwi. If you assume in bad faith that I was point scoring instead of educating you and other readers, that says more about you than me. Accept in good faith that someone may simply be trying to teach you something. Akld guy (talk) 21:45, 22 August 2017 (UTC)
 * I really could care less, Akld, about the politically correct neologisms of a nation half the size of NYC. The Gould standard is gold enough for me. Go file a complaint at WP:ANENGVAR. μηδείς (talk) 03:33, 23 August 2017 (UTC)