Wikipedia:Reference desk/Archives/Science/2015 December 11

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.

December 11

Calculating orbital distance based on 2 observations

[1]

The two observations show movement of the new dim object relative to Alpha Centauri A and B. So, why can't we find the movement, say in terms of arc seconds, then figure out how long it would take to move 360 degrees based on that, and derive the orbital distance from that ? Some potential problems I see:

1) This assumes it's in orbit about the Sun. If it's in the Alpha Centauri system, or not bound to either, then this won't work. (If in the Alpha Centauri system, it might have already made an orbit in the observation period, but then presumably we would have more observations).

2) This won't take acceleration into account.

3) Let's assume it's in our solar system. This doesn't consider movement towards and away from Earth (which is essentially the same as towards or away from the Sun, at that distance). Would movement toward or away from the center of our solar system change the calcs ? (I realize that this implies acceleration, but if that accel is only towards and away from the center, rather than normal to this direction, does that matter ?)

Also, a second Q: There are questions about why it wouldn't have been observed more often. Could it be varying in signal strength for some reason, in the frequency detected (short microwave wavelengths), such that it drops below the observation threshold ? StuRat (talk) 15:21, 11 December 2015 (UTC)

• The original paper is here. The authors aren't even sure that it is orbiting us – all they can say is that it's comoving with the Centuari system but doesn't seem to be bound to it. As for why it wasn't seen before, they suspect that it "could not have been detected with the VLT-NACO data [the last major survey of that area of space, done in 2004 to 2005] because, due to the intense glare from Cen AB, it was intentionally blocked out of view." This makes precovery very difficult - if you look at Figure 2, between about 2000 and 2010, it was almost overlapping the Centauri system. It's only now that it's moved further away from the star system that it's become visible. Smurrayinchester 15:53, 11 December 2015 (UTC)

(edit conflict) Let us assume that object is in orbit around the sun and that we as observers are roughly in the focal point of the ellips (i.e. comparatively close to the Solar System's center of mass). Then the orbit is decribed by Kepler's second law: $P\cdot {\tfrac {1}{2}}r^{2}{\frac {d\theta }{dt}}=\pi ab$ with P the period, r the object's current distance from the focal point (us) and a and b the semi-minor and semi-major axes of the ellipse. From the two pictures, the only thing we can directly calculate is ${\frac {d\theta }{dt}}$, so there are simply too many unknowns. Only if we assume that the orbit is circular (i.e. $r^{2}=ab$) can we find the orbital period, and, from that, the object's distance. - Lindert (talk) 16:04, 11 December 2015 (UTC)

Thanks. I take that the mean that movement (and hence acceleration) towards and away from the Sun does affect the orbital period. StuRat (talk) 16:08, 11 December 2015 (UTC)

Yes, you could say that; of course if you knew the acceleration, you'd also know the distance simply from Newton's law of gravity. - Lindert (talk) 16:34, 11 December 2015 (UTC)

My guess is that to predict where something is going to be in the future - you need to know where it is now, and it's velocity. Each of those requires three numbers (like X,Y,Z and dX, dY, dZ) - so you have a system of equations with six unknowns. If your observation is only a 2D thing (azimuth/elevation for example) then each observation provides only two "knowns" - so you need three observations to produce six knowns with which to calculate the six unknowns. SteveBaker (talk) 18:20, 11 December 2015 (UTC)

Yes, that is fundamentally true, but I doubt that three measurements is always sufficient. And for orbiting objects that are very far away, some of the information may be redundant, considering the accuracy of the measurements, because we pretty much know already that they will move in a 'straight' line across the sky, hence the third measurement may give us only one extra unknown. - Lindert (talk) 18:40, 11 December 2015 (UTC)

Sure - and you can start making wild assumptions, like assuming that an unknown object is probably perfectly orbiting in the plane of the ecliptic and probably not moving in a retrograde orbit. That sort of assumption might turn out to be completely wrong, but it's a good way to constrain the unknowns from 6DOF to lower degrees of freedom. That lets you construct an assumed trajectory even if you don't have enough information to completely constrain it.

If you read astronomy journals about Near-Earth object detection, you'll see that these types of assumptions are frequently applied when data is very sparse. As more observations become available, there is confirmation or refutation of previous assumptions; and the error-bars just get tighter and tighter after each subsequent measurement. The math of Bayesian linear regression provides one way to combine each observation. The more data you have, the less you need to depend on wild assumptions.

In actual reality, the model needs to be bigger than 6DOF, because orbits are imperfect; objects wobble; other solar-system objects affect the orbit too.

I subscribe to Orbital Debris Quarterly (a publication of NASA Johnson Space Center); every few months they give an update of latest news and events in orbital debris science. In addition to recent observations, they also publish a software catalog - including ORDEM, an algorithm and a whole digital database. Here's information on how they take the measurements. You can see how they go from a first-observation to the construction of an orbit ephemeris. A corresponding object catalog, the NEO catalog, tracks objects that are less likely to reenter Earth's atmosphere as debris.

Nimur (talk) 20:06, 11 December 2015 (UTC)

Thanks for the replies so far. Anyone care to take on the second part ? StuRat (talk) 07:24, 12 December 2015 (UTC)

Not much to add, except that an ideal orbit has six orbital parameters (six degrees of freedom), which can be expressed in various ways, see orbital elements. A still photograph gets you two pieces of information so it takes at least three such photographs to solve for the parameters/orbital elements. Bubba73 ^{You talkin' to me?} 05:50, 14 December 2015 (UTC)

Sure, I'll take a shot at the second part. Yes, it's possible, but I don't think it's likely. Red dwarfs and brown dwarfs exhibit two main forms of variability: they could be BY Draconis variables or flare stars, or both. The former generally vary by less than 0.5 magnitudes. Flare stars can increase in brightness by a few magnitudes in just a few minutes, and fade to normal in a few hours. So it's possible that both observations of this object happened right when it was flaring, whereas other surveys photographed it when it wasn't flaring. This would be rather extraordinary luck, but it isn't impossible. --Bowlhover (talk) 09:28, 14 December 2015 (UTC)

Super attacks

Is it possible for a person to have 2 or more independent (I.e. 1 didn't cause the other) life threatening attacks at the same time? For example a heart attack and a stroke or an anaphylactic shock and a seizure etc. What would happen in such cases? — Preceding unsigned comment added by 2A02:C7D:B901:CC00:D198:4CE1:9927:97A3 (talk) 21:22, 11 December 2015 (UTC)

Let's say that one is taking care of bees, and has a brain attack. He falls down, his protective clothing is opened and he is stung by a bee. This produces anaphylactic shock. The scenario is plausible. I won't speculate about the outcome. Tgeorgescu (talk) 00:07, 12 December 2015 (UTC)

Although you could argue those weren't truly independent. However I think gets on to an important issue, namely that it seems unlikely you can come up with a clear definition of independence that will be able to seperate all cases. For example, the OP said "I.e. 1 didn't cause the other", but this seems a very weird definition of independence. While definitely when one clearly causes the other, e.g. if someone has a heart attack and crashes their car suffering major traumatic injuries you would probably say these aren't independent; there would also be cases where you have simultaneous life threatening problems where it's questionable if one cause the other, but which most people would say aren't independent because they both happened because of something else which arguably wasn't life threatening as and of itself. Consider for example if you fall down because you screw up and in doing so badly damage your spine and also get stung by a bee or other examples where the fall wasn't cause by a life-treatening "attack" but causes two or more life threatening "attacks" (you could likewise imagine the car crash scenario). In fact, you get the added complication of what's life threatening. Consider the case of [2] for example. A Paradoxical embolism would normally be consider a very serious or life-threatening condition, but the reason would be because it could easily cause a stroke, MI or that sort. See also [3]. And while doctors would normally do their best to figure out what's going on, it may not always be the case that they can be sure whether the heart attack caused a clot which cause the stroke [4] or something else happened. (And besides, even when they do believe this is what happened, it would often be at most an educated guess based on symptoms etc. Particularly if they were successful in treatment so there was no autopsy. Nil Einne (talk) 06:54, 12 December 2015 (UTC)

And, of course, it is possible that a person could have multiple independent acute conditions at once. You could compute the probability if you knew the portion of the person's recent life during which they had an acute attack of each type. For example, if they have acute attack type A 1% of the time, and acute attack type B 2% of the time, and they really are completely independent of one another, then (1/100) x (2/100) of the time, or 2/10000ths or 1/5000th of the time, you would predict they would suffer both concurrently. However, this is such a small portion of the time that if there is even the slightest dependence, either positive or negative, that will swamp out this number quite quickly. StuRat (talk) 07:19, 12 December 2015 (UTC)

Doctor: "Mr Burns, you seem to have every disease known to man, simultaneously, such that the effects of each is precisely countered and balanced by the effects of the others."

Montgomery Burns: "So, I'm in perfect health then !" StuRat (talk) 07:28, 12 December 2015 (UTC)

Two questions about steroids and testes

I would like to know if the following rumors is correct or no. people say about taking steroids as a bodybuilder, causes to the testes to be smaller. Is it right? Second, if it's right, it doesn't make sense to me - apparently, because that testosterone (if I understand well that is the steroid that the build-builders usually use) supposed to make it bigger as a male hormone. 22:10, 11 December 2015 (UTC) — Preceding unsigned comment added by 92.249.70.153 (talk)

Confirmed in Anabolic steroid#Feminization. It says it inhibits natural production of testosterone, and sperm production. Graeme Bartlett (talk) 22:22, 11 December 2015 (UTC)

As for why, "Feminization can be noticed in men due to the conversion of excessive testosterone into female hormone estrogen." [5]. StuRat (talk) 07:20, 12 December 2015 (UTC)

Hmm, could that be a separate effect, though? I knew about the conversion, but I thought that was the explanation for breast hypertrophy, not shrunken testicles.

The latter I've always attributed to negative feedback — the body senses that it has enough testosterone so it stops making it. But maybe I just assumed that. Actually the linked article doesn't seem to say one way or the other. Anyone have more information on this? --Trovatore (talk) 08:02, 12 December 2015 (UTC)

Yes, that seems reasonable, but we need a source to confirm. StuRat (talk) 06:47, 13 December 2015 (UTC)

There is another factor that might contribute. Bodybuilders are by definition (usually) causing their muscles to become bigger, and as a result they look bigger overall; however, if their testicles remain at best the same size (having no significant muscle mass) these will look smaller by comparison. {The poster formerly known as 87.81.230.195} 185.74.232.130 (talk) 14:10, 14 December 2015 (UTC)

For further reading, see Anabolic steroids – a problem in popular sports by Ulrich Hoffmann, Institute of Pharmacology, Ernst-Moritz-Arndt University. Alansplodge (talk) 17:34, 14 December 2015 (UTC)

Here is my understanding. If a male takes in testosterone from an outside source, that "tricks" the body into thinking that it has enough testosterone. Therefore, the body "thinks" that it does not have to produce any more testosterone (since it senses enough there already). Thus, the testes do not have to do their "job" anymore. Thus, they shrink. That's my understanding. Thanks. 2602:252:D13:6D70:14DE:69F5:F4C:EAE3 (talk) 00:01, 16 December 2015 (UTC)