Wikipedia:Reference desk/Archives/Science/2019 May 13

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.

May 13

PC1 vs PC3

Does anyone know of a way to plot principal component 1 vs principal component 3 in R?

I need clarification, My guess is this is related to vector math, where PC1 and PC3 are two perpendicular vectors and R is the resultant of the two. Is that what you meant ? SinisterLefty (talk) 01:15, 13 May 2019 (UTC)

I think this is a request for assistance using the plotting tools of the R (programming language).

Here's a blog on computing and visualizing PCA in R, from a popular website for R users.

Here is a basic and an advanced plotting tutorial from a math professor at the University of Georgia.

Nimur (talk) 02:28, 13 May 2019 (UTC)

Yes, that's what I meant!

If you mean Nimur described what you meant, then he has the answer for you. If you want to know about the general answer, not specifically in the R programming language, then we would need to know how R is specified, and whether it's a fixed or floating vector. SinisterLefty (talk) 08:20, 13 May 2019 (UTC)

Sailing into the sun in a submarine made of ice

Alright, I just heard of superionic ice. [1] Made at 25000 atm of pressure, melts at 8500 K sorry!. Which is more than anything else I've heard of, and more importantly, the color temperature of my computer monitor. Consists of mobile hydrogen ions moving through a lattice of oxygen ions, which I suppose makes it a cross of solid and liquid plasma? By now I'm officially intrigued.

Question is, what can you do with this stuff? It can be made in a diamond anvil, but I don't know what happened to the anvil afterward. So I figure the safest thing you can do with a snowball is throw it into the Sun, where the pressure might rapidly rise to a comfortable level and the temperature might not be exceedingly high. You'd have to wrap it up in something first (possibly not a diamond anvil) to keep it safe under the harsh conditions of Earth and space, and perhaps have some interface inside to more normal matter and temperatures if you want to have passengers (sturdier than mambo chickens, though) to plant the flag of life in the Sun. So... could something like this be stably linked to ice VII, which in turn can be stably held in diamond inclusions, or is there some range of temperatures where neither form of ice is stable? Can a submarine in the sun tap effectively into all those juicy magnetic field lines for enough current to run a hella air conditioner? Could it communicate back results by any effective channel, hard X-rays or something, or would it have to wait to be shot out in a prominence before preparing its swan song? Has anyone had fun speculating with this tech? Wnt (talk) 00:59, 13 May 2019 (UTC)

As our article states: "If it were present on the surface of the Earth, superionic ice would rapidly decompress", so the same would be true anywhere lacking the extreme pressure required to maintain it. If you could build a pressure vessel capable of maintaining the required pressure, it would melt as it approached the Sun. Thus your "flying it into the Sun" scenario wouldn't work. SinisterLefty (talk) 08:36, 13 May 2019 (UTC)

Use an asteroid or something. Or borrow a planet from the Earthlings, since they're going to destroy it anyway. Wnt (talk) 11:57, 13 May 2019 (UTC)

An asteroid or terrestrial planet wouldn't work. Our article states that ice giants, like Neptune or Uranus, are required. SinisterLefty (talk) 13:39, 13 May 2019 (UTC)

Problem is, before plunging into the sun you first encounter Roche limit, where self-gravity cannot maintain pressure, so this doesnt work either. Gem fr (talk) 14:34, 13 May 2019 (UTC)

You can use a material on the outside that will evaporate fast so that it will yield the necessary pressure. Count Iblis (talk) 22:09, 13 May 2019 (UTC)

I apologize -- I saw it was 8500 K, checked the NYT article and it seemed to say it was so (and the paper was paywalled)... but NYT was reporting Fahrenheit, not Kelvin! So probably there's nothing very special about the melting temperature of superionic ice compared to all the usual not-quite-going-to-work-for-a-sun-shot substances like tungsten. The photosphere goes below this temperature in places, but I suspect the pressure is too low there... though I still kind of wonder from this if any kind of superionic oxygen "snow" could precipitate out of the sun's plasma, were the oxygen ions to be prone to stick together to the exclusion of electrons. Wnt (talk) 01:43, 14 May 2019 (UTC)

Bergsonian Time

According to Cybernetics: Or Control and Communication in the Animal and the Machine, in the first chapter "Newtonian and Bergsonian Time", Norbert Wiener, considering the Second Law of Thermodynamics is asserting : "Tidal forces between the planets introduce a degree of decay over cosmological time spans, so that Newtonian mechanics does not precisely apply". Searching for "limitations to Newtonian mechanics" I find this about the Limitations of Classical Mechanics, in which Newton’s Universal Law of Gravitation no longer accurately models the observed universe and needs to be replaced by general relativity. Is Norbert Wiener alluding to general relativity? Such a description for a source of decay is confusing me, I'm used to suppose that tidal forces may have been Isaac Newton's initial source of inspiration. Or is the laws of thermodynamics in any manner contradicting his Principia ? --Askedonty (talk) 10:52, 13 May 2019 (UTC)

Laws of thermodynamics do not, in any manner, contradict Newton Principia. These fully account for tidal forces

I don't understand this quote from Wiener, and wouldn't consult Wiener on such a topic anymore than I would Einstein for medical advise.

Gem fr (talk) 11:48, 13 May 2019 (UTC)

• "The Principia" can refer to a lot of things: a model for planetary motion, or laws from which it can be deduced (Newton's law of universal gravitation + Newton's laws of motion). The model for planetary motion of Newton (well, Kepler already had the result based on experimental data) is a really good model, but it has limitations (as all models have). One limitation is the existence of forces other than gravitation; those cause the initial model (which assumes gravity is the only force at play) to be wrong; however, the effect of those forces can be described by Newton's laws of motion.

Tidal forces are the reason the moon moves away from the Earth (at about 4 cm/year). Looking at the balance of energy, the gravitational potential of the Moon/Earth system is converted into thermal energy by the friction of tides on Earth.

Relativity shenanigans are another source of discrepancy that affect all Newtonian mechanics, but they have little to do with tidal forces. Tigraan^{Click here to contact me} 12:26, 13 May 2019 (UTC)

I did a google on "relativity mercury orbit". Is that what we are talking about? 85.76.141.245 (talk) 17:15, 13 May 2019 (UTC)

Yes. Tigraan^{Click here to contact me} 10:53, 14 May 2019 (UTC)

I don't think it referred to general relativity but rather that tidal forces suck up some rotational energy from planets with atmospheres or oceans (think of the earth's oceans sloshing around because of the lunar tide) that turns into heat that gets radiated into space. 67.164.113.165 (talk) 02:45, 14 May 2019 (UTC)

Thanks a lot. Thanks particularly Tigraan, Gem fr. Your detailed answers insured my view enough for bringing back to my mind the path to the correct explanation (not to the question that was asked, but to why it had to look like it needed to): "weak field approximations and post-Newtonian expansions in powers of v/c—will, under certain conditions, overlap" (International Journal of Theoretical Physics, Ivan Tolstoy 2001,2002). Post-Newtonian expansions "abound" starting from 1850 approximately, and at the turn of the century. Wiener relies heavily on Bergson's philosophy in the next of his book too, so I guess that's all of it. --Askedonty (talk) 08:43, 15 May 2019 (UTC)

How was the size of blood unit set?

How was it decided how large the units we store blood in should be? Presumably people who need transfusions don't reliably need the same amount of blood each time, so the unit size could have been almost anything.

I guess we need to balance a few factors, such as not having the unit too large so we often waste some which isn't needed and/or it's not safe for people to donate that much. But it feels like we could have picked an awful lot of sizes, and the ideal size might be based on the most common causes of needing blood which may change over time. Is the unit size ever reassessed? 51.9.121.241 (talk) 19:35, 13 May 2019 (UTC)

A unit of blood is ~525 ml, roughly equivalent to 1 pint, and represents about 10% of the blood in the average healthy adult (8–12 pints), according to this. Mikenorton (talk) 19:48, 13 May 2019 (UTC)

Our blood donation article however, gives 450 ml as the normal size of a blood donation and say that it is equivalent to a pint. It makes sense to store and use it in the same units that it's donated. The limit to how much a single donor can safely provide was presumably discovered by trial and (potentially fatal) error. Mikenorton (talk) 20:00, 13 May 2019 (UTC)

Matters seem somewhat more complicated AND simpler to my non-MD eyes.

more complicated: Looking at articles Packed_red_blood_cells#Single_unit_transfusion, blood transfusion, Single unit transfusion, Fresh frozen plasma and related, it appears that quite a number of different volume -- 50 to 650 ml -- exist (including smaller for children!), some of them explicitly mentioning that they are made out of a single 450 ml blood donation, but not all.

simpler: Obviouly (and explicitly in the french wiki) a «unit» is just a plastic bag holding the product, whatever content it has, pretty much like a pill for other medecine.

I gave blood quite a number of time, and quite obviously the system do not allow accuracy of the volume given. Would more or less half a liter, enough to be sure that you can process it to get 450 ml of blood product

But then again, I am no MD.

Gem fr (talk) 21:29, 13 May 2019 (UTC)

450mg of water weighs about 1 pound, so I'd guess they measured it by weight rather than volume originally. Dmcq (talk) 21:04, 13 May 2019 (UTC)

Scratch that - I'd go with it being US pints. Dmcq (talk) 21:10, 13 May 2019 (UTC)

Hence the rhyme school children used to be taught "A pint is a pound the world around" (ignoring the fact that no one in the world except the U.S. uses pints or pounds). The general implication is that a pint of water (or mostly-water-based liquids) should weigh a pound of weight. --Jayron32 12:02, 15 May 2019 (UTC)

Here in the UK, I still use pints and pounds, but my British Imperial pint weighs 20 ounces (a pound and a quarter). To anyone who says that Britiain is entirely metric, I point out that I always buy milk by the pint. Dbfirs 17:41, 15 May 2019 (UTC)

If anyone thinks the UK went entirely metric, they should hang around horse races, and they won't think that furlong. SinisterLefty (talk) 17:50, 15 May 2019 (UTC)

Potentially relevant: Hypovolemia § Stages shows "stages" of blood loss based on volume. Up to around 750 ml blood loss in a healthy adult leads to only minor symptoms. Obviously, in whole blood donation, the donor can only donate a safe amount at one time; similarly, in transfusion, a healthy blood volume needs to be restored, while avoiding excess (hypervolemia). --47.146.63.87 (talk) 23:39, 13 May 2019 (UTC)

One factor which would tend to make them want to use smaller units of blood is that if they only give a partial unit to a patient, the rest is presumably wasted. Thus, to prevent waste, they should have units so small that there's no need to give partial units. Of course, there may be other factors tending to encourage larger units. SinisterLefty (talk) 17:33, 15 May 2019 (UTC)