Episode 110 - Solar System Mysteries "Solved" by PseudoScience, Part 2 - The Pioneer Anomaly
Recap: Imagine, if you will, that NASA sent two spacecraft out into the solar system, launched in 1972 and 1973. Their sexy encounters with Jupiter and Saturn over a mere two and six years later, respectively, they were alone and pretty much forgotten by most of the world that had watched their launch, though they continued to return data to interested scientists. It was only when something weird was observed, first as early as the 1980s, but not seriously investigated until 1994, that the world began to again take notice, and as is often the case of something unexplained, everyones' pet idea came out of the wood work.
Puzzler for Episode 110: There was no Puzzler in this episode.
Q&A: This episode's question comes from Brandon from Georgia, USA, who asked: My question is regarding solar conditions. How is it the surface temperature is colder than the core, or the corona. I am fine with the reduction in [temperature] from core to surface, but how does it heat back up again leaving the surface all the way to the corona?
To answer this question, I need to backup a bit for those listeners who don't know what we're talking about.
The sun, our nearest star - unless you count a cloaked Nemesis that emits no light, no gravity, no heat, no nothin' except shows up in a few amateur photos and videos every now-and-then - is divided into several different layers internally. And these aren't arbitrary divisions just for the sake of dividing things up, completely different things happen in these layers.
In the center is the core, which is defined as the only area where nuclear fusion can take place because that's the only region where it's hot enough and the pressures are high enough for it to happen. It's about 20% the radius of the sun and gets up to nearly 16 million Kelvins. I'll be covering that more in a future episode about the Iron Sun.
Out from the core is the radiative zone, an area where it's still really hot and really dense, but fusion can't occur, and energy simply streams outwards. The temperature here drops from 7 million Kelvins at the outer core to 2 million at the outer radiative zone, which is out to about 70% the way to the surface.
Out from that is the convective zone, which is where the sun's material actually swirls up and down like giant conveyor belts, much like Earth's mantle. This is a much more efficient way to move heat, and the temperature drops from 2 million Kelvins to the surface temperature of the star, about 5700 Kelvins.
Which gets us to the photosphere, the effective surface. The photosphere is not a surface that you can walk on - it's not solid - but it's where the sun is finally low enough density that it is no longer opaque to visible light, so the light streams off from it. Hence "photo" = light, "sphere" = ... sphere. It is the "light sphere" that we observe as the sun.
But, the sun doesn't stop there, it has what can be thought of as an extended atmosphere. And, that's where it gets weird. The atmosphere, which is only visible if you block out the sun - and before space telescopes, it was only visible during total solar eclipses - has five layers itself: The temperature minimum, chromosphere, transition region, corona, and heliosphere. The heliosphere is what extends from the sun out until it meets the heliospheres from other stars -- the Voyager spacecraft are starting to pass out of the sun's helisphere.
The temperature minimum region is about 500 km thick, and is as cool as 4100 K, cool enough for simple molecules to exist like carbon monoxide or even water. Above that is the chromosphere, which comes from the Greek word "chroma" for color because it's visible as a flash of color just before and just after totality during a total solar eclipse. It's about 2000 km thick, about the size of the continental US if I'm using very round numbers, and it starts to increase in temperature from the minimum to 20,000 K -- practically 3.5 times hotter than the photosphere. Then there's a 200 km transition region where the temperature increases to 1 million K, which is the base of the corona, which reaches 1-2 million K, but it gets as hot as 8-20 million K.
And herein lies Brandon's question: How the heck does it get so hot?! (my paraphrasing)
When I was in college, just a bit ago, there was no answer. This was just one of those mysteries. Sure, there were a lot of ideas, things like acoustic waves, fast and slow magneto-acoustic body or surface waves, Alfvèn waves, current dissipation, microflares, mass/particle flows and magnetic flux emergence, magnetic carpets, magnetic reconnection ... the list goes on.
This is one of the many reasons why we have launched solar observatories into space, to try to figure out some of these mysteries. One thing it has going for it is that the corona is very thin, it has very little mass. So, if you have the same heat input to, say, a giant pot of water versus a giant pot of air, you are going to much more quickly heat the air than the water because there's so much less mass there to heat. So, even though it's outside the opaque part of the sun and is larger in volume, it doesn't require that much energy -- only about 1/40,000th the amount of light energy that actually escapes the sun.
But, we still don't really have a theory that's well tested and is better than all the others. At the moment, the two leading candidate hypotheses are wave heating and magnetic reconnection. The planned NASA mission Solar Probe +, now planned for 2018, is intended to get within 10 solar radii of the sun and test these models.
In recent years, the magnetic reconnection idea has been gaining traction, the very basic and simplified idea being that magnetic fields loop out from the sun and create electric currents in the corona. When they suddenly collapse, they release all their built up energy as heat and wave energy, heating the corona.
I don't want to get more into that explanation because this is very far from my field of expertise, but I've provided several links in the shownotes for interested parties. But, bringing it all back, the answer is that we don't really yet have an answer. Some good ideas that seem to work, but none that has been shown conclusively to be the case.
- Wikipedia: Pioneer 10 || Pioneer 11 || Pioneer Anomaly || Spin-Stabilization || Radiation Pressure
- arXiv.org: Turshev, S.G. et al. (2012). "Support for the Thermal Origin of the Pioneer Anomaly."
- arXiv.org: Submitted papers about the Pioneer Anomaly
- Technology Review: Pioneer Anomaly Solved by 1970s Computer Graphics Technique
- Planetary Society: Pioneer Anomaly Solved!
- The Space Review: Review of "The Pioneer Detectives: Did a Distant Spacecraft Prove Einstein and Newton Wrong?"
- Q&A References
- Logical Fallacies / Critical Thinking Terms addressed in this episode: Argument from Ignorance
- Relevant Posts on my "Exposing PseudoAstronomy" Blog
Claim: Imagine, if you will, that NASA sent two spacecraft out into the solar system, launched in 1972 and 1973. Their sexy encounters with Jupiter and Saturn over a mere two and six years later, respectively, they were alone and pretty much forgotten by most of the world that had watched their launch, though they continued to return data to interested scientists. It was only when something weird was observed, first as early as the 1980s, but not seriously investigated until 1994, that the world began to again take notice, and as is often the case of something unexplained, everyones' pet idea came out of the wood work.
The "something weird" was where the spacecraft actually were in space. Scientists tried to figure out every single force acting on the spacecraft - from the gravity of all major bodies to how the very few particles they encountered in interplanetary space would affect the spacecraft by acting as a drag force.
But, what the scientists observed was that the Pioneer spacecraft were thousands of kilometers closer to the sun than they should be. That sounds like a lot, and on the scale of Earth, that's a lot. But, it's tiny relative to the distances that Pioneer 10 and 11 had travelled so far and for so long.
In Episode 108, I gave some practical examples of measurement uncertainty. These were all factored in. As was the possibility of a software glitch. And each new, independent calculation showed what the old calculations did, that the Pioneer craft were not where they were supposed to be.
The force acting on the spacecraft must have been tiny -- it was only (8.74±1.33) x 10^-10 m/s^2. For most of you who have no idea what this number means, that's about 100 BILLION times less than the force of gravity at Earth's surface. Incredibly tiny. But, over time, it adds up.
And, it was still adding up. As in, when the spacecraft were first observed to be in a different location, or at a different distance anyway, than they were supposed to be, and then measured later, they were MORE off from where they were supposed to be. It wasn't a one-off thing, whatever was affecting the Pioneer craft was still affecting them, and it was still affecting them at about the same amount as a function of time.
I know what you're thinking -- "What about the Voyager spacecraft?" The Voyagers did not show this effect, but they couldn't; it certainly would have been lost in the noise, below the level of measurement accuracy.
The reason is that the Pioneer spacecraft were spin-stabilized, meaning that the spacecraft physically spins, using gyroscopic principles to stay oriented the way we wanted it to. Because of that, it was much easier to predict EXACTLY how the physical environment - like high-speed particles from the sun called the "solar wind" - would affect the spacecraft.
Voygers were not. They only had thrusters to orient themselves the way they needed to be. Because thruster firings were common on the Voyager craft, tiny, unpredictable changes happened, and these were larger than the tiny acceleration that the Pioneer craft showed.
Other spacecraft that we've sent out have shown inconclusive results, again due to the level of measurement uncertainty. So, it's really the Pioneer craft that are the only things - for now - that we have to go on.
For many years, we simply didn't know. There were many mainstream ideas, but there were many non-mainstream. As is typical when we don't know why something is the way it is, Argument from Ignorance kicks in and the crazies come out with their own "theories."
There were some more mainstream ideas that still may fall under the category of "out there," though not impossible. Papers were published with titles like, "Scalar Field Models: From the Pioneer Anomaly to Astrophysical Constraints." Or, "Pioneer Anomaly? Gravitational Pull Due to the Kuiper Belt." Another was, "Discrete Fields and the Pioneer Anomalous Acceleration." Perhaps a more esoteric title was, "A mirror world explanation for the Pioneer spacecraft anomalies?" which proposed that hypothetical "mirror matter" could account for the anomaly.
Perhaps the largest or one that had the most support was MOND, the abbr. for "Modified Newtonian Dynamics." As I discussed in episode 108, gravity is the worst-measured of the four fundamental forces in the Standard Model of physics. And, we know that Newton's formulation is not 100% accurate because of Relativity. The MOND proponents suggested that the Pioneer spacecraft were so far away that what they offered us was now a much larger baseline against which to measure the force of gravity. Like trying to estimate when you will arrive at a location only a few seconds after you left versus an hour after you left -- you have a much longer period of time and distance and so can get a better estimate.
In MOND's case, the tenant is that Newton's formulation for gravity works because the terms in his equations dominate everything else. But, there is a tiny addition sign in there that he didn't know about, but it's so small of an effect that it can only be measured when you measure over huge distances and or huge masses. Incidentally, during the 1990s, MOND was the main competing model against dark matter, but dark matter won out as the explanation for why the stars on the outskirts of galaxies were orbiting as fast as they were. But, MOND was proposed as an explanation for the Pioneer anomaly during its height, before the dark matter theory had won out over it in cosmology.
Another model was simply errors in observations or recording of data. But, that was ruled out when numerous completely independent analyses all came to the same conclusion. Another win for peer review and independent replication.
A third explanation was clock acceleration. This follows the same premise as MOND, that our current theories break when you get to the very large. In this case, if the universe is indeed expanding, then there is an increase in the background gravitational potential, which based on relativity, in a changing gravity field, time changes. In this case, the team advocating this model said that the clock frequency onboard the Pioneers - which was needed because they had to tell us what time they radioed into Earth so we could calculate how long it took the signal to get back - would change only by 1.5 Hz in 8 years. A very tiny amount that could fit into their model, though more and more data meant it got more and more complicated to explain the actual observations.
All of these were actually possible. Unlikely, but possible. Turns out that they weren't the case, but they were, well, possible.
And then you get to the pseudoscience, the explanations that had already been shown to be wrong for other reasons, but re-appropriated by their proponents to explain the Pioneer anomaly. The main two that I've seen are Planet X and the Electric Universe.
Planet X was invoked simply as the accelerative force that slowed the craft down, making them be closer to the sun than they should have been. The problem with this is that it's easily calculable how far away the planet would need to be, or how massive it would need to be -- you have to assume one or the other to do the calculation -- and the planet would have to be visible and affect other objects in the solar system in obvious ways ... and it didn't. Also, the acclerative force was reasonably constant, but it should have changed if it were a planet as the planet orbits the sun over many years.
Electric Universe is something that I have not yet addressed on this podcast, though I'll dip my toes into it in an upcoming episode on the claims of James McCanney. The basic premise of EU is that everything in the universe has an electric charge and runs off of electricity. The sun doesn't shine through fusion, comets don't shine because they sublimate and reflect sunlight, planets move through an electric field and it does stuff ... it's all very complicated but has a rather large following online.
One of the proponents of EU, Wallace Thornhill, wrote:
"After launch, a spacecraft accepts electrons from the surrounding space plasma until the craft's voltage is sufficient to repel further electrons. Near Earth it is known that a spacecraft may attain a negative potential of several tens of thousands of volts relative to its surroundings. So, in interplanetary space, the spacecraft becomes a charged object moving in the Sun's weak electric field. Being negatively charged, it will experience an infinitesimal "tug" toward the positively charged Sun. Of most significance is the fact that the voltage gradient, that is the electric field, throughout interplanetary space remains constant. In other words, the retarding force on the spacecraft will not diminish with distance from the Sun. This effect distinguishes the electrical model from all others because all known force laws diminish with distance. The effect is real and it will have a fundamental impact on cosmology and spacecraft navigation because Pioneer 10 has confirmed the electrical model of Stars!"
If that sounds like a lot of gobbley-gook, it is. More on why in future episodes.
For the moment though, perhaps it's easiest to make a comparison with episode 69, The Solar Neutrino "Problem." In that episode, the claim was that we didn't find the right amount of neutrinos from the sun to have its energy explained by fusion for a star its age. Therefore GodDidIt 6000 years ago. But, rather than give up science and turn to religion, this anomaly led us to a completely new model of neutrinos, that they have mass, and therefore can change type, and when we observed the right number of neutrinos across all types to account for fusion, the neutrino "problem" was solved, and another God of the Gaps was closed.
What's Going On
That's what happened for the Pioneer anomaly, as well, and in 2012, the problem was solved: Heat.
To understand this, we have to go back to how the spacecraft were constructed. They were unique among other missions, even a spin-stabilized craft like New Horizons which is on its way to Pluto. The craft were both made so that the antenna needed to talk with Earth was on one side, and pretty much all the science instruments were on the other side. Same as with other spacecraft, but this was spin-stabilized.
What makes it different from New Horizons is that the power source - the radioisotope thermoelectric generators or "RTGs" - were way far away from the spacecraft body, and connected via struts and cables to the main spacecraft. This means that ALL the unknown and complicated heat patterns from the RTGs on the oddly shaped spacecraft can pretty much be ignored. Unlike New Horizons, where the RTGs are on the body of the spacecraft. As in, people have suggested we look at New Horizons to see if it experiences the same effect, but it would be much harder to do.
So, while the actual theoretical explanation was suggested back in 1998, we didn't have the telemetry records of the spacecraft temperature, nor could we create a detailed thermal - or heat - model of the craft with the 1998 technology. And the thermal models at the time said the effect should decrease, which hadn't been observed.
Over the next 14 years, these issues were solved. Telemetry records were found, so we now knew how hot some parts of the spacecraft were and how much power they were using. Thermal models were built inside rapidly better computer programs, and these models could be tested against the telemetry data.
Many groups converged on the same answer, but it's a 2012 paper by Slava Turyshev and others that's most often cited. The term is "Thermal Recoil Force."
The only way the spacecraft can lose heat is by radiating photons, just like the sun, only much, much colder. Since the instruments are fed power by the RTGs, and they are all on one side of the spacecraft, this radiation is anisotropic, meaning that it's not uniform in direction. Most is going to be radiated away from the spacecraft from the instruments on one side, effectively in the direction opposite the antenna. And, since the antenna is facing Earth, that means these photons are being radiated more away from Earth than towards it.
And, photons have momentum. Meaning that they can push with a certain force. This was first theorized by Kepler back in 1619 as an explanation for why a comet's tail always faces away from the sun, even though half the time the comet is moving away from the sun, meaning it's chasing its tail.
Because photons have momentum, and thus exert a force, we call the phenomenon "radiation pressure." And, just like thrusters, this emitting of photons preferentially from one side of the Pioneer spacecraft is going to give you an equal and opposite force, as per Newton's Third Law of Motion.
Putting it all together, in simpler terms: The spacecraft is slightly hotter on one side, it emits light from that side, which gives it a tiny thrust in the opposite direction, towards Earth. The reason we needed that telemetry and those precise thermal models is because the exact locations of where it was hotter will vary the exact anisotropic direction of the heat.
When all the models were done and published, they agreed with each other and predicted the Pioneer anomaly to about 80%, at least the Turyshev model was to 80%. Then, they analyzed all other sources of uncertainty and possible amounts of measurement error. When they plot their prediction and a 1-sigma uncertainty, and plot the data and its 1-sigma uncertainty, the two 1-sigma uncertainty envelopes overlap. That means that even though their model was only 80% the observed value, with the uncertainty factored in, it was more like 80±30% versus 100±40%. Since they overlap, we're good.
And keep in mind that 1-sigma uncertainty is the same as about a 70% confidence level, meaning that if we were to make the observation another 100 times, about 70% of those times, what we observe would be in that range.
This is one of those rare cases where I recommend reading the paper, and I've linked to a free version in the shownotes. It's 5 pages long, and the abstract is very readable, as are the conclusions. It's a good example of how science is done.
I've put off doing this episode for a long time because there are a lot of sacred cows in it and I wanted to make sure that I at least tried to do it some justice. Hopefully I've done that.
I like the tale because it's another example of the triumph of science over pseudoscience, in contrast with what very well may have been a new physics or revised physics.
We had an observation, one on an experiment the likes of which we had never done. The precepts of science were followed by many and various explanations were offered. Some involved modifications to the laws of physics as we knew them, but they were ones that at least worked in concert with and not against previous observations. For example, even though MOND would change how we thought of gravity, it still could easily account for every single previous observation and experiment of gravity.
That's in contrast with the pseudoscientific explanations, like Planet X and the Electric Universe, which had been debunked for years if not decades, but still had and have ardent believers who were desperate to incorporate some legitimate science into their model. "Look, look!" they cried, "The Pioneer anomaly fits right in with exactly what our model predicts!" Unfortunately, no calculations or actual predictions were included, or they were included very, very rarely. Again, in contrast with legitimate scientific pursuits.
And then came the "Big Reveal," something that happened within the lifetime of this podcast. I couldn't have done this episode in my first year of podcasting.
It turned out that known effects, when actually modeled accurately, could completely account for what was observed, at least within the measurement uncertainty. Even though the actual explanation may sound like techno-babble, it's not, and it has effects and evidence completely outside the Pioneer anomaly. And, just like other examples, such as the solar neutrino "problem" offered up by young-Earth creationists years ago as proof positive that they were correct, the pseudoscientists weren't.
Provide Your Comments:
No comments have been provided.