Wikipedia:Reference desk/Archives/Science/2017 March 8

Science desk
< March 7	<< Feb | March | Apr >>	March 9 >

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.

March 8

Year Span of 485-490 Days

Are there any known planets out there with a year span of 485-490 days ? I ask this because the number of days in a Julian cycle is 4×3651⁄4 = 1461 = 3×487. — 79.113.199.220 (talk) 00:51, 8 March 2017 (UTC)

Do you mean our Earth days of 24 hours or another planet's number of rotations during its orbit around the sun?--Aspro (talk) 00:59, 8 March 2017 (UTC)

Initially, I had Earth days in mind, but, now that you've mentioned it, I would be curious about the latter as well. — 79.113.199.220 (talk) 18:47, 8 March 2017 (UTC)

Not in our solar system. Mars already has 667 days in a year. If I understand how it works, you would need a planet between Earth and Mars to have a year with a length between those two. StuRat (talk) 01:01, 8 March 2017 (UTC)

If you look at List of exoplanets and sort by orbital period, you will see that none are on that list. There is a lot of selection bias going on here, and planets with short orbital periods are much more likely to be identified. Graeme Bartlett (talk) 01:13, 8 March 2017 (UTC)

I was looking at the exact same table, while you typed that! So, "HD 95127 b" is 482 days, and "HD 564 b" is 494 days. For sure, there will be lots - billions upon billions! - which we're yet to discover.86.20.193.222 (talk) 02:14, 8 March 2017 (UTC)

The length of a year is not exactly 365.25 days. So, whatever mystical calculation you plan on doing will be wrong. We have a leap year every four years EXCEPT years ending in 00 unless the year is divisible by 400. 209.149.113.5 (talk) 14:35, 8 March 2017 (UTC)

For starters, my calculations (mystical, or otherwise) have never been wrong so far. :-) Secondly, I would be more interested in whether this value has ever been used historically in either astronomical or calendrical computations. — 79.113.199.220 (talk) 18:47, 8 March 2017 (UTC)

Yes. The length of a year has been used in many calculations along with many things, such as the number of books in the Old Testament or the number of planets in the Solar System or the number of toes on a four-toed hedgehog. Pick any day of any year and it is very likely that someone made a calculation that something would happen on that day. 209.149.113.5 (talk) 21:53, 8 March 2017 (UTC)

(Are you joking, or do you genuinely not know the difference between astronomy and astrology ?) — 79.113.199.220 (talk) 02:30, 9 March 2017 (UTC)

Also, tropical or sidereal? Or anomalistic? Sagittarian Milky Way (talk) 15:27, 8 March 2017 (UTC)

Both. If either you or someone else know of any planet out there whose `year span` (tropical, sidereal, anomalistic, etc.) lasts anywhere in between 485 and 490 days, please let me know. — 79.113.199.220 (talk) 18:47, 8 March 2017 (UTC)

Falling asleep in classrooms

In one of the lectures of a community college classroom, I dozed off and then rested during a brief video lecture. When the video ended, I just suddenly regained consciousness (or woke up). Why is that? --George Ho (talk) 02:46, 8 March 2017 (UTC)

Wikipedia's only major coverage seems to be Sleep#Awakening, but that has some references which may lead you to sources to answer your question. --Jayron32 02:54, 8 March 2017 (UTC)

I was a champion lecture-sleeper in college, and it always seemed that sudden changes woke me up. If nothing else, everyone getting up at the end o the lecture would do it. I once dozed off during a lecture on sleep, and became an object lesson for the classroom. I don't have an answer for you specifically, but apparently the nature of arousal from sleep is a deeply complicated field [1]. Someguy1221 (talk) 03:55, 8 March 2017 (UTC)

OR: There can also be a very subtle trigger you become accustomed to. I regularly woke up at exactly 5:59 each morning, and my alarm was set for 6. I eventually figured out that the alarm made a little click when it hit 5:59, and that woke me up, not because it was itself loud enough to wake me, but because my brain took it as a warning to get up now as a jarring alarm would go off soon. StuRat (talk) 04:46, 8 March 2017 (UTC)

I quite often wake up just before my alarm goes off but not sure there is any clicking going on. One weird thing when I used to use a radio alarm is I used to dream about a particular song. Then I would wake up due to the alarm and that song would be playing. I remember hearing the alarm switching on waking me up and the song in my sleep definitely happened before the alarm went off. Explain that! Polyamorph (talk) 20:16, 8 March 2017 (UTC)

For the first one, assuming there is nothing else which happens before the alarm which you've learnt to associate with the waking up, the obvious issue is how consistent the time of your alarm is. If you always wake up at the same time, or at least the same time on the same day then you may very well have learnt to wake up at that time. If the times are not consistent at all, it's more complicated. First you need to really check (as StuRat mentioned) there is nothing else happening that you're associating with the alarm. I would generally expect a mobile phone won't for example, but you'd still need to check somehow. Then you'd need to check your are really waking up before most of the time and not just a few times and remembering those times, the timing of 'just before' is also relevant. Unless you've kept a careful diary (which is complicated anyway), there is a strong risk of confirmation bias. In fact, depending on the circumstances you may very well be waking up more than that and not remembering.

I admit I've sometimes felt I did the same, and with a mobile phone, and the times weren't consistent at all but I suspect it's probably a combination of confirmation bias and that I tend to wake up a lot anyway. I may also be taking clues from stuff like how dark it is etc. A good activity tracker which detects how many times you're waking etc would help here although you have to be very careful about timing and data quality etc, especially if you're trying to use it to determine if you really wake up before.

For the second one, two likely explainations are your memory of hearing the alarm switch on waking you up is faulty or you didn't actually have a song in your sleep before the alarm went off. People can often be a bit confused when waking up and can still be in a semi-dream state, so any memory of what happened and which one was first is always a bit circumspect.

Nil Einne (talk) 22:55, 8 March 2017 (UTC)

It is also common for the brain to subjectively move and expand dream memories in time. For instance, I have sometimes been awoken by external noises but had the memory to have seen the corresponding reason of that noise occur a while before the noise actually happened and woke me up (as if my brain had foreseen it). Similarly, we can have the impression to have had a long dream, while it may actually have been produced in a very short moment. I am not sure if this is related to the same phenomena, but hypnosis is notable as being able to produce false memories which are perceived by the subject to have occurred back in time, but were actually forged by suggestion during sessions (and hence hypnosis being considered a misleading investigation tool, other than aspects of it being considered pseudoscience)... PaleoNeonate (talk) 00:14, 9 March 2017 (UTC)

That's interesting! That a sleepy brain will mix up the order of short term memories which are perceived as real time experiences. Thanks for your answers.Polyamorph (talk) 09:28, 9 March 2017 (UTC)

I've often noticed that, when I flinch from a sudden noise, my memory of the flinch precedes my memory of the bang. From which I infer that between sensation and memory are multiple neural paths, differing in length. —Tamfang (talk) 06:27, 12 March 2017 (UTC)

More OR: I have found that when I'm dozing, like half-asleep, my hearing gets more acute, i.e. small noises which wouldn't normally grab my attention might wake me up. ←Baseball Bugs ^{What's up, Doc?} carrots→ 05:30, 8 March 2017 (UTC)

See also the appendix of this paper. Count Iblis (talk) 08:17, 8 March 2017 (UTC)

How tall was Homo erectus?

On the Homo erectus talk page there was a question about this paper:[2] The page doesn't appear to have any information at all about how tall and how heavy an adult homo erectus was. Does anyone have a source for that information? --Guy Macon (talk) 06:36, 8 March 2017 (UTC)

Apparently researchers in the field are reluctant to give height estimates. This paper suggests a height of between 1.64m and 1.68m for a particular H. erectus individual, and then immediately casts doubt on it. This source, which I can only see the preview of, states that one individual was estimated would have reached 1.85m in height had it survived (a reference I believe to Turkana Boy, whose potential height estimate was later lowered dramatically), and that no adult H. erectus skeleton has been recovered with an estimated height less than 1.58m. So in terms of prehistoric hominids, it sounds like they were actually quite tall, though they have also been described as quite skinny. There are uncertainties in the height measurements because no complete skeleton has ever been recovered of this species, and there are inherent uncertainties such as the distance between vertebra. Someguy1221 (talk) 07:30, 8 March 2017 (UTC)

I suspect that this is related to the fact that everyone has heard of T. Rex but nobody seems to have heard about Sarcosuchus imperator or Deinosuchus riograndensis. In the popular culture, big prehistoric reptilian predators all went extinct. Having one that has close relatives still alive today doesn't fit the narrative. Likewise, in the popular culture recent hominids were big, muscular brutes. Having one that was tall, skinny, walked upright and had a small brain doesn't fit the narrative. --Guy Macon (talk) 05:22, 9 March 2017 (UTC)

Wind noise

Two questions:

1) How exactly does wind translate into noise ?

2) I've noticed that below a certain speed, winds don't seem to make any noise at all. Is this due to laminar flow versus turbulent flow ? StuRat (talk) 16:22, 8 March 2017 (UTC)

• This looks like some good recent research on the general topic, as it pertains to automobiles. This paper discusses the process as it pertains to sound engineers and microphones specifically, both seem to discuss wind noise in general, as well as how it pertains to specific fields of study. --Jayron32 16:28, 8 March 2017 (UTC)

• Also relevant to consider is that there are two, distinct, definitions of sound: the physics definition of a mechanical wave and the psychological definition of sound as qualia. When one asks "what causes wind noise", that implies both "how does wind cause vibrations in the air" and "how does the human mind perceive such noise". --Jayron32 16:33, 8 March 2017 (UTC)

Thanks so far. Anyone else ? StuRat (talk) 18:13, 11 March 2017 (UTC)

Why are headlights green?

I'm not even sure "headlights" is the right term. I see these during the day. Newer cars in some cases have a row of white dots (almost blue, actually) which might be a half circle. Or they might be a full white circle in other cases. I started wearing a new pair of sunglasses and these lights are the only thing they change for white to a yellowish green.— Vchimpanzee • talk • contributions • 18:15, 8 March 2017 (UTC)

Daytime running lamps? Sagittarian Milky Way (talk) 18:22, 8 March 2017 (UTC)

Can you include a picture? It makes it easier for us to understand what you're talking about. --Jayron32 18:23, 8 March 2017 (UTC)

This isn't exactly the same as what I've been seeing but it's close enough.— Vchimpanzee • talk • contributions • 19:36, 8 March 2017 (UTC)

These LED lights tend to be a bit bluer (to me, at least) than 'normal' headlights. Colour perception is notoriously difference between people, maybe you're perceiving them as greener. Fgf10 (talk) 19:49, 8 March 2017 (UTC)

They do look blue to me. But not with these sunglasses on. Older sunglasses that broke didn't do this.— Vchimpanzee • talk • contributions • 20:45, 8 March 2017 (UTC)

Lots of sunglasses are designed to block blue light (the hardest color to focus, I think?). So from your description, it seems likely that the (bio-adjusted) intensity profile of the lights has some red, a medium amount of green, and peaks in the blue. Your sunglasses filter out the blue, leaving the green. I don't have enough information to give a more precise answer, and in any case color vision is very complex so any treatment like this is bound to be oversimplified, but my speculation is that this is what is going on. --Trovatore (talk) 20:52, 8 March 2017 (UTC)

Although it is possible to make a white LED, most such LEDs are actually clusters of LEDs emitting three colors that combine to appear white. If you block one completely, the effect would be quite noticeable. Someguy1221 (talk) 01:33, 9 March 2017 (UTC)

I don't know who mentioned blocking anything "completely". Many sunglasses are designed to block blue light, but of course not completely. --Trovatore (talk) 02:04, 9 March 2017 (UTC)

I mentioned it, to make a point, which I guess didn't come through. I have a suspicion that applying a narrow absorption spectrum to a "peaky" emission of light (google "white LED emission spectrum" to see how bimodal these are) will have a more dramatic effect than applying the same peak to a blackbody spectrum that appears to have the same color. Someguy1221 (talk) 03:28, 9 March 2017 (UTC)

Indeed. I'd note there's no particular reason you can block one LED unless it has some special polarisation or something. What you will block are wave lengths and it doesn't matter how they are produced. Certain methods will produce more peaky spectra but it's too complicated to say just RGB will do that. Nil Einne (talk) 02:40, 9 March 2017 (UTC)

I thought it would have been obvious I was referring to a sunglass coating whose emission/reflection spectrum is centered on one wavelength that happens to be close to one of the peaks in the LED's emissions. Someguy1221 (talk) 03:30, 9 March 2017 (UTC)

The problem is this is still IMO conflating two different things.

TLDR version is if phosphor coated LEDs had not come to predominate because of luminous efficacy advantages, I'm not convinced typical white LEDs which were multichip would necessarily be more peaky than they are now. (The SPD may be different sure.) Whether this would be from more than 3 LED colours or some other strategy I can't say but it seems to me some strategy would have been used. In other words, the issue of white LEDs generally being RGB other than being untrue is IMO mostly a red herring even if the peakiness of the SPD is relevant. Multichip LED don't have to be and probably wouldn't be so peaky if they were the norm and phosphor coated LEDs are still a bit peaky and would potentially be more so if no one cared about them.

Looking online, it does seem most RGB LEDs still have a fairly peaky SPD. One of the best I found was this using a collimator [3] but even that is still tripeaked. Much more common is something like this [4].

But low CRI phosphor white LEDs can also be peaky particularly in the blue part of the spectrum (which seems particularly relevant to this question), see [5] and [6] or the earlier sources. More to the point, I don't see any reason why a RGB LED intrinsicly has to be tripeaked. Making a fairly broad SPD R, G and B LED is probably difficult and may require a phosphor coating or similar, probably one reason development hasn't been particularly successfully. And I see even less reason why a phosphor coated LED couldn't be tri-peaked (and I also question if tri-peaked matters so much here since we're probably only talking about cutting of one peak). Possibly the development of phosphor coated LEDs was helped by earlier research for other purposes e.g. fluorescent lights but I'm not sure (I may have known at one time, but can't remember).

Regardless even if it is easier, I think there's a fair chance a key reason why decently broad SPD and decent CRI phosphor white LEDs are so much better than RGB LEDs is since that's where the research is due to the luminous efficacy advantage. IIRC green LED luminous efficacy is the biggest problem and until this is solved, RGB LEDs are only going to have niche applications.

And some of these applications are reducing since the research into phosphor white LEDs means some degree of spectrum tunability is now achieved using multichip LEDs with a white LED as I mentioned (see e.g. [7]). Besides that I also came across some sources mentioning using more than 3 colours [8] [file.scirp.org/pdf/CS20120100016_16408165.pdf] (one seems commercial, one research), I'm not sure whether these are at all common but ultimately most manufacturers are going to concentrate on what works and is cost effective etc, not in proving they can create a broad SPD RGB LED.

While broad SPD may be related, it isn't necessarily the target anyway. CRI is generally more important although the usefulness of CRI to LED colour rendering performance is often questioned for LEDs anyway [9] [10] [11] [12] [13]. Not that this means such a peaky SPD is considered good, in fact IIRC (and also what I think I saw from a quick skim through the sources) one of the concerns is that peaky spectrum can achieve a decent CRI but even there a different peaky spectrum may often be better yet achieve a lower CRI. But this also doesn't mean the typical idea is just need a broad SPD with no peaks in the visible spectrum. And this is complicated by other targets like luminous efficacy, CCT etc too. E.g. I didn't look that well into this research [14], but it sounds like they're suggesting using low CRI lighting in unimportant places. (Although 2007 is very old in LED research.)

Typical white LEDs is perhaps not the best reference point anyway. There is a fair amount of research into how different SPD, CCT etc affect headlamp performance. This is mostly in how it affects illumination and visibility of stuff that matters to drivers at night but I'm sure there must be some into how it affects performance of daytime running lights. Which isn't to say there isn't also a lot of marketting etc. [15] [16].

Nil Einne (talk) 13:57, 9 March 2017 (UTC)

That isn't true. Most white LEDs are blue LEDs with a phosphor coating. RGB LEDs are only used for niche cases. If you don't believe me look at a white LED or do a search. Many LEDs do use clusters but that's a different point. Note that clusters of different colours are sometimes used for improved colour rendering or adjustment but these aren't necessarily be RGB. They may in fact have one or more white LEDs. Nil Einne (talk)`

News to me! And you're right! From sleuthing on Google Scholar, it looks like single-color LED with a phosphor coating exceeded three-color LEDs in efficiency sometime in the mid-late 90s. Someguy1221 (talk) 03:28, 9 March 2017 (UTC)

No problem and sorry for the tone. I was tired and also editing from a mobile device. Nil Einne (talk) 10:21, 9 March 2017 (UTC)

Ah, no offense taken. My fault for relying on an electronics textbook written in '93. Someguy1221 (talk) 23:01, 9 March 2017 (UTC)

I wore different sunglasses today and the lights are back to white. I bought the sunglasses yesterday since the new ones that turn the lights green were the last pair I had left that weren't broken in some way. And they specifically said they block ultraviolet light. I don't know what I did with the tag on the others.— Vchimpanzee • talk • contributions • 17:00, 10 March 2017 (UTC)

Could they have been high-intensity discharge lamps ? "Beginning in the early 1990s, HID lamps have seen applications in automotive headlamps. Xenon, or high-intensity discharge (HID), lighting provides brighter headlights and increases visibility of many peripheral objects (e.g. street signs and pedestrians) left in the shadows by standard halogen lighting." These tend to look blue, versus the more yellow look of traditional headlights: [17]. StuRat (talk) 21:54, 10 March 2017 (UTC)

I think we established what I was seeing with the new sunglasses weren't those. And this week I quit wearing the sunglasses that turned those lights green, saving them in case something happens to the newer ones.— Vchimpanzee • talk • contributions • 18:13, 11 March 2017 (UTC)