Reflections on Pioneer

March 31, 2011 7:35 AM

As they leave the solar system, the Pioneer spacecraft have anomalously decelerated, pointing to a possible gap in our understanding of gravity. Now, a computer graphics technique known as Phong shading predicts that the Pioneer anomaly is just a side effect of how the shape of the spacecraft reflects sunlight.
posted by jjray (57 comments total)

Tractor beam
posted by The Whelk at 7:40 AM on March 31, 2011 [2 favorites]

This application of Phong shading is okay, I suppose, but I'm wondering if we couldn't put the technology to more profitable use predicting the future paths of mirrored balls being juggled by a blocky man standing on an endless checkerboard.
posted by No-sword at 7:47 AM on March 31, 2011 [43 favorites]

It was originally developed to handle the reflections of visible light from 3D objects but it works just as well for infrared light

So would this also hold true for pretty much all electromagnetic radiation, such as radar etc?
posted by Old'n'Busted at 7:49 AM on March 31, 2011

I don't know much about spaceflight, but I know something about computer graphics.

If I understand this correctly, the acceleration of Pioneer is affected by the radiation of the sun, just like a solar sail. The direction of reflection of the photons off of the spacecraft affect the direction of the force applied to the object. Is this paper saying that the previous models of that effect never considered directional reflection?
posted by demiurge at 7:50 AM on March 31, 2011

I've never been comfortable with the rush to build new physics on a handful of real-world anomalies derived from very sketchy observations noted merely as a side of the actual experiment. There is a reason that a non-zero number of physics jokes start off with assuming a cow is a sphere.

By all means, eliminate more sources of error and actually test this with some nearly-spherical probes, but until then, can we please make those MOND people shoo?
posted by adipocere at 7:51 AM on March 31, 2011 [1 favorite]

So would this also hold true for pretty much all electromagnetic radiation, such as radar etc?
I don't think Phong shading does diffraction - so no, when the wavelength approaches the same kind of physical scale as the object (or components of it) it presumably won't work quite right.
posted by edd at 7:53 AM on March 31, 2011

Happy to help
posted by phong3d at 7:54 AM on March 31, 2011 [30 favorites]

Photonic pressure can be counterintuitive. I study the mechanical properties of single stem cells by shining unfocused infrared laser beams at them, which makes them deform from the photonic load alone. The technique is called optical stretching. The story goes that the PI that conceived the idea told his graduate students to go shine lasers at a cell to ostensibly "squeeze" it by photon reflection. The students came back and said, "It's not squeezing the cells, it's stretching them." It turns out that as the photons go through the cell, moving from a lower to a higher refractive index, the cell surface gets a kick backwards from the momentum transfer. As the photons leave the cell, the surface gets a kick forwards, and the cell ends up stretching. It's all understood now, but it can be tricky to intuit the answer before correctly modeling the details. Let's not be hard on the scientists and engineers who tried to generate as many possible reasons for the anomaly.
posted by Mapes at 7:59 AM on March 31, 2011 [12 favorites]

How in the world are they measuring such minuscule differences in velocity at such fantastic distances?

Here I thought I was being nice and precise out in the shop: Measure twice cut once "yep, that's about halfway to the next 32nd mark..."
posted by notyou at 8:01 AM on March 31, 2011 [2 favorites]

My reading of the link says that previous work on the problem did not model the reflection of the radiation as accurately. They basically ignored 2nd order effects. Even without the careful modeling though they were able to explain 67% of the problem. Using the better light reflection routine now fixes the rest of the problem.

Stunning though that people were willing to make new physical laws based on 33% error in a difficult calculation.
posted by Chekhovian at 8:03 AM on March 31, 2011 [1 favorite]

If I understand this correctly, the acceleration of Pioneer is affected by the radiation of the sun, just like a solar sail

No, infrared light produced by the nuclear reactor power supply bounces off the back of the dish, resulting in slowdown.
posted by delmoi at 8:07 AM on March 31, 2011 [1 favorite]

Radiation does some weird things on sufficiently complex surfaces. I was at a symposium where the speaker off-handedly showed us a pair of surfaces for which the Casimir force pushed them apart.
posted by adipocere at 8:09 AM on March 31, 2011 [1 favorite]

How in the world are they measuring such minuscule differences in velocity at such fantastic distances?

Primarily because the probes have had over 40 years for small differences in velocity to add up to big ones.

And modern astronomical instruments are crazy accurate in measuring the doppler effect. Although is the wikipedia article correct in referring to it as a blueshift?
posted by KirkJobSluder at 8:14 AM on March 31, 2011

There is a reason that a non-zero number of physics jokes start off with assuming a cow is a sphere.

Ridiculous. A cow is a torus, like most digesting creatures.
posted by hippybear at 8:16 AM on March 31, 2011 [40 favorites]

No, Taurus is the bull. My charts show that the cow is an Acquarius.

*lights incense*
posted by the quidnunc kid at 8:19 AM on March 31, 2011 [9 favorites]

Also, does the Pioneer anomaly explain why I can't spell?
posted by the quidnunc kid at 8:25 AM on March 31, 2011

This is the lowing of the cow of Aquarius...
posted by hippybear at 8:27 AM on March 31, 2011 [7 favorites]

Much simpler rendering techniques can explain E1M8, the Phobos Anomaly.
posted by kersplunk at 8:29 AM on March 31, 2011 [8 favorites]

When the moo is in the seventh house...
posted by hippybear at 8:30 AM on March 31, 2011 [9 favorites]

hmmm...didn't think you could make a sailboat go with a fan on board... ;)

How in the world are they measuring such minuscule differences in velocity at such fantastic distances?
oh, those NASA kids...they pretty much LIVE for precice...saw a bit in a documentary program where they were demonstrating a machine that tested the roundness of large ball bearings...they put the ball bearing in the machine, tested it, amplified its flaws by a million and overlaid an image of it on a circle...it was off by a twinge, but pretty much still a circle. then they put a billiard cue ball in it, did the same thing. it was a triangle. then they pulled out a particle counter (looked like a tiny kitchen scale) and dropped a pinch of sand on it...didn't miss a single grain...then they did it with flour.
but that's nada compared to the tricks they can do with light (and i'm imagining they're doing these velocity measurements with some kind of radio doppler effect)...you've heard maybe about all the extra-solar planets they've been finding? (over 500 to date) ...most of those have been found using optical measurements of the velocity of stars many, many quadrillions of miles away with an accuracy of ~3m/s (a slow jog) to find out how much their planets are pulling them around (not much, but enough)...
you want to be proud of being human? be proud of optics.
posted by sexyrobot at 8:31 AM on March 31, 2011 [7 favorites]

The direction of reflection

The angle of the dangle affects the direction of reflection.
posted by backseatpilot at 8:31 AM on March 31, 2011 [3 favorites]

delmoi: So it's a reverse solar sail-like effect due to the radiation emitted by the spacecraft? That seems easier to overlook.
posted by demiurge at 8:36 AM on March 31, 2011

Actually I did it with my mind. I was stretching my newly developed telekinetic abilities.

Sorry. My bad.
posted by Splunge at 8:42 AM on March 31, 2011 [1 favorite]

I've never been comfortable with the rush to build new physics on a handful of real-world anomalies derived from very sketchy observations noted merely as a side of the actual experiment.

Ignoring the "sketchy observations" bit as non-applicable, not only am I completely comfortable with this I'm very uncomfortable when it isn't done. When you have grand theories, all anomalies are going to be handful-sized. And what on earth does it matter that the observation was a side thing? As Asimov said, "The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' (I found it!) but 'That's funny ...'"
posted by DU at 8:42 AM on March 31, 2011 [4 favorites]

Ignoring the "sketchy observations" bit as non-applicable, not only am I completely comfortable with this I'm very uncomfortable when it isn't done. When you have grand theories, all anomalies are going to be handful-sized. And what on earth does it matter that the observation was a side thing? As Asimov said, "The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' (I found it!) but 'That's funny ...'"

We're just scratching the surface of the universe beyond Earth, and we're understandably on the look out for things that make break holes in our favorite theories, or help us to draft crazy new hypothesis! That's what's so exciting about it!

...but the media coverage of this stuff does encourage one to become jaded and cynical, they really do leap on the most extravagant mumblings and rumors and blow them wildly out of proportion.
posted by Stagger Lee at 8:48 AM on March 31, 2011

DU, I'll trot out the Sagan "... but they also laughed at Bozo the Clown."

There's a lot of things that appear funny in the lab. So, devise experiments explicitly designed to test The Funny. Here we have a case where The Funny led to people, rather than thoroughly going through known science and engineering (1970s computer graphics!) or coming up with a good test, just stampeding towards new physical law.

People have been continually casting about for fifth and sixth forces since long before I was an undergrad. Just casually searching for one, I came across one from 1986.

The fainter the additional phenomenon is, the harder we ought to look for second, third, and forth order effects in known science. Such was the case here.
posted by adipocere at 8:59 AM on March 31, 2011

> the Pioneer anomaly is just a side effect of how the shape of the spacecraft reflects sunlight infrared light emitted from the spacecraft.
posted by brenton at 9:11 AM on March 31, 2011

So that's where the old guy in Reboot got his name.
posted by EndsOfInvention at 9:17 AM on March 31, 2011 [2 favorites]

If I understand this correctly, the acceleration of Pioneer is affected by the radiation of the sun, just like a solar sail. The direction of reflection of the photons off of the spacecraft affect the direction of the force applied to the object. Is this paper saying that the previous models of that effect never considered directional reflection?

From what I can tell, it's saying that the heat of the power generator on the other side of the reflector is generating infrared light that is bouncing off the back of the reflector and pushing it back towards the sun.
posted by empath at 9:21 AM on March 31, 2011

Here we have a case where The Funny led to people, rather than thoroughly going through known science and engineering (1970s computer graphics!) or coming up with a good test, just stampeding towards new physical law.

I really don't think this is what happened. The original guys surely did a million things before they even announced anything for fear of looking stupid. And many fine minds applied many existing theories to try to figure it out.

One or two came up with relatively minor tweaks of existing physical laws¹. They made predictions with these tweaks and found they didn't tally (with Pluto, for instance).

If the "1970s computer graphics" was so obvious, why didn't you come up with it?

And really, what does it matter if there's a "stampede towards new physical law"? If those laws work, they'll be kept. If they don't, they won't. That's what happened here. The system worked.

¹The most earthshatteringly massive of which was "maybe at really really really far distances theres a tiny correction factor". This is a lot less of a change in physical laws than Einstein proposed and on the basis of a lot more, and a lot stronger, evidence.
posted by DU at 9:45 AM on March 31, 2011

We've had spacecraft leave the solar system before, so why haven't we seen this happen with, say, voyager?
posted by Krazor at 10:10 AM on March 31, 2011

There's a lot of things that appear funny in the lab. So, devise experiments explicitly designed to test The Funny. Here we have a case where The Funny led to people, rather than thoroughly going through known science and engineering (1970s computer graphics!) or coming up with a good test, just stampeding towards new physical law.

Well, there are a lot of things that were discovered that way. Generally speaking, if your formulas have super minor exceptions, they're wrong. Physical laws should work perfectly when you control for everything and if they don't there is something else involved. Our understanding of quantum physics came about because of minor anomalies in what was predicted by Newtonian physics.
posted by delmoi at 10:20 AM on March 31, 2011

We've had spacecraft leave the solar system before, so why haven't we seen this happen with, say, voyager?

Answered in the first link in the FPP:

The Voyagers flew a mission profile similar to the Pioneers, but were not spin stabilized. Instead, they required frequent firings of their thrusters for attitude control to stay aligned with Earth. Spacecraft like the Voyagers acquire small and unpredictable changes in speed as a side effect of the frequent attitude control firings. This 'noise' makes it impractical to measure small accelerations such as the Pioneer effect; accelerations as small as 10−9 m/s2 would be undetectable.[16]

posted by hippybear at 10:21 AM on March 31, 2011 [1 favorite]

We've had spacecraft leave the solar system before, so why haven't we seen this happen with, say, voyager?

Klingons?
posted by Herodios at 10:22 AM on March 31, 2011 [1 favorite]

Somehow that's really anticlimactic and really cool at the same time.

Does Phong shading really give an accurate representation of the reflections from a surface? I thought it was a clever way to emulate realistic lighting, but not necessarily physically correct. (I would have assumed there were known formulae to calculate reflections exactly, but that they were just too complex to use for realtime graphics.)
posted by lucidium at 10:39 AM on March 31, 2011

If the "1970s computer graphics" was so obvious, why didn't you come up with it?

When people got all hot and bothered about this originally, I said "You watch. That's a complicated shape with a lot of radiation impinging on it and being generated by it. I don't buy it [new forces]." I have been skeptical all along.

The burden of proof is on people who want to introduce new physics, not on people who do not buy in. Hang out in a physics department sometime — you would not believe the stuff that arrives from people who are convinced that if we just accepted everything they said, we'd see that electrons were indeed shaped like pyramids (actual example). Not tetrahedrons, pyramids. Square bottoms and all. Tiny Cheopses flying about. If we would just do the math for them, we'd see.

As with perpetual motion machines, free energy, and most other things, I stick to "Sounds great. Lemme know when it works." Statistically, me not bothering really pays off.

If you want me to bother, cool. You're writing the grant, right?
posted by adipocere at 10:50 AM on March 31, 2011 [1 favorite]

One thing that folks may be overlooking is that scientists always come up with a lot of bad ideas — by which I mean that anyone, even on a good day, will spend most of their time on hypotheses that turn out to be false.

When some bit of anomalous data comes up, you start brainstorming for explanations. Maybe you generate a thousand hypotheses; you throw away 900 after a few minutes of thought, and you find obvious holes in the next 90 after a week or two, and then you reject 9 based on experiments or simulations, and then... — Well, frankly, then a lot of the time the thousandth hypothesis fails too, and you go back to the drawing board, or wait for someone else to think of one that never occurred to you.

There's also a tendency for scientists to focus their work on theories they don't like very much. This seems paradoxical, but really it's a pretty natural way to operate. In general, if we can think of two possibilities — one that's straightforward, one that's horrible and disruptive — our impulse is to rule out the disruptive one first. (Doctor to patient: "Well, it could be brain cancer, but more likely it's just a migraine. Still, we're going to test for brain cancer first, just to be sure.") The hypothesis that everything we know about gravity is wrong is a pretty damn disruptive one. So some people have rushed to work on it — not because they believe it, necessarily, but perhaps just because they're hoping to disprove the damn thing and get it out of the way.

Publish-or-perish also puts pressure in the same direction, for what it's worth. Disproving some crazy hypothesis is still good for a publication — but it's probably going to be easier than disproving one that's more plausible. And on the other hand, if you end up confirming the crazy hypothesis, it might just make you famous, so it's win/win. We call this "picking the low-hanging fruit first."

To non-scientists, this can look like some sort of crazed panic. "My God! One tiny little anomalous data point, and they just start spewing out these wacky new theories left and right! And they go running straight for the most implausible ones! Why get all worked up like that over a fraction-of-a-percent error?" When in fact, nobody's getting worked up at all. It's just business as normal, and they're doing what scientists always do — floating all the explanations they can think of, and then ruthlessly culling the ones that don't pan out, starting with the most far-fetched or disruptive ones.
posted by nebulawindphone at 10:53 AM on March 31, 2011

The burden of proof is on people who want to introduce new physics, not on people who do not buy in.

And this is what happened. Everyone proposing a theory has attempted to prove it, these guys included. I don't understand the complaint.

Was even one single textbook altered with "new physics" because of this problem? (Other than a statement of the problem/?)
posted by DU at 10:54 AM on March 31, 2011

Though on non-preview, the folks doing the complaining appear to be in physics themselves, and probably didn't need my amateur sociology of science schtick. Sorry if that sounded condescending. I've had a lot of "Hurr hurr you research nerds think you're so clever" conversations lately, and I may be just a wee bit jumpy about defending the cause.
posted by nebulawindphone at 10:57 AM on March 31, 2011

Strangely enough a cow is roughly a torus. The hole running from the mouth to the anus turns it (and us) into a doughnut shape.
posted by zengargoyle at 11:08 AM on March 31, 2011

cow is roughly a torus

By A. Elk
posted by Trochanter at 11:13 AM on March 31, 2011

I thought MOND was originally developed as an alternative to exotic dark matter to explain a general problem of galaxy rotation? Certainly MOND has been falsified there as well, but it's not as it was invented entirely because of the Pioneer anomaly.
posted by KirkJobSluder at 11:52 AM on March 31, 2011

Oh good God, they've hit the side of the petri-dish.
posted by Slackermagee at 12:09 PM on March 31, 2011 [2 favorites]

I don't buy using an approximation like Phong shading to explain such a tiny anomaly.

Why didn't they use Ray Tracing or even Photon Tracing for precise results? Both methods are computationally tractable with modern computers. Phong shading was invented because computers in the 1970s were too slow for more accurate models of light propagation.
posted by monotreme at 12:59 PM on March 31, 2011

I don't buy using an approximation like Phong shading to explain such a tiny anomaly.

Well, according to the last link in the FPP,

In particular, Phong shading has allowed the Portuguese team to include for the first time the effect of heat emitted from a part of the spacecraft called the main equipment compartment. It turns out that heat from the back wall of this compartment is reflected from the back of the spacecraft's antenna (see diagram above).

Now, I don't know a lot about Phong shading, but perhaps there's something in how the parameters are submitted for processing, or whatever, which allowed for this calculation to be easily included in the results. That's my best guess. Perhaps it could also have been done with Ray or Photon tracing, but maybe there's something inherent in how Phong shading is manipulated which made it an ideal interface for testing out the hypothesis.
posted by hippybear at 1:40 PM on March 31, 2011

A quick search reveals that neither ray tracing or photon tracing have an inherent reflection model. Instead, they use other algorithms, including Phong, to estimate the effect of surface reflections. And of course, the phenomenon of interest here is momentum transfer across the surface of the dish, so Phong is a reasonable method of estimation. Tracing is useful if you want to know what happens to photons after hitting the dish. But we already know that most will be reflected into interstellar space.
posted by KirkJobSluder at 2:03 PM on March 31, 2011

Phong is one of the worst reflectance models, though, and has been long supplanted by better ones (although it is still in heavy use in real-time graphics because it is computationally cheap). In particular, Phong is not a physically based model-- it's meant to look good, not reproduce the physics of reflection.
posted by Pyry at 2:15 PM on March 31, 2011

That's useful information, thank you.
posted by KirkJobSluder at 2:47 PM on March 31, 2011

I'm still sticking to my theory from 7 years ago.
posted by luvcraft at 3:31 PM on March 31, 2011

Stunning though that people were willing to make new physical laws based on 33% error in a difficult calculation.

I think it's awesome that smart people were willing to say "You know what? We might be missing something important, and we could also be dead wrong about something we've believed for a long time. It's not very likely, but let's find out for sure."

If everything I've learned from cartoons about history is right, the alternative goes something like:

"...well, that's odd."
"STONE HIM!"
posted by obiwanwasabi at 4:00 PM on March 31, 2011

In particular, Phong is not a physically based model-- it's meant to look good, not reproduce the physics of reflection.

I was just going to post about this but couldn't recall whether Phong was physically based or not. Since it isn't, what's the basis for using it in the paper? Or is there some other reason it's applicable?
posted by howlingmonkey at 4:18 PM on March 31, 2011

Didn't notice that the paper was linked in the article:

This method provides a simple and straightforward way of modeling the various components of reflection, as well as a more accurate accounting of the thermal radiation exchanges between the surfaces on the Pioneer spacecraft.

They don't seem to address that specific point in the paper. It seems strange that they would be using a non-physical model without better justification.
posted by howlingmonkey at 4:25 PM on March 31, 2011

I blame the Tholians.
posted by bwg at 6:20 PM on March 31, 2011

This fairly-long, well-done PopSci article (Dec. 2010) includes background and recent cud-chewings from Turyshev and Toth.

they themselves have become convinced of the thermal cause of the anomaly

This 5MB PDF from 2004 is a technical synopsis with lots of figures.
posted by Twang at 6:47 PM on March 31, 2011

Ridiculous. A cow is a torus, like most digesting creatures.

I ♥ your maths.
posted by eriko at 8:43 PM on March 31, 2011

If this holds up, it's a killer awesome result of science.

"WOW UNKNOWN FORCE"

"Unknown by people playing lame 3D games, dude...."

Nice work.
posted by eriko at 8:45 PM on March 31, 2011

Science Fuck Ya!
posted by Mitheral at 9:41 PM on April 9, 2011