Exoplanets - Suspicious evidence and fantastical conclusions

In summary, Paul thinks there are some strange claims being made about exoplanets and that the evidence is not conclusive. He also mentions that the pool of knowledge gathered over the last two decades makes it more likely that the findings are correct.
  • #1
PaulCam
5
0
Hi,

I thought I would come and ask here as something has been bugging me for quite a while.

Exoplanets.

I'm not being bugged by the fact we are looking for them or that they exist. I am a bit suspicious of some of the evidence and how it's interpreted. While the adage is "Extraordinary claims require extraordinary evidence" I feel that some of the exoplanet findings are the other way around. It seems to me that the evidence being found amounts to extraordinary and the conclusions gleaned from this evidence are extraordinary.

Take for example gas giants that have orbital periods of only a few days. Another example is the higher than expected number of very close and very large planets. The astrophysics community seem to be going all out to find explanations for these extra-ordinary examples.

Is it not more likely that there are issues with the evidence gathering? Is it not scientific that if an experiment results in completely unexpected or extraordinary results that you first question your experimental approach to double and triple check it is not an artefact of your experiemental design... before you start coming up with fantastical conclusions like a gas giant with an orbital period of a few days... which does not seem likely!

I think the science is too young and the methods used to vague and speculative, open to too much interpretation to be making the kind of claims that are a weekly occurrence in the popular science media.

Am I the only one?

Paul
 
Astronomy news on Phys.org
  • #2
While some scientists jump too quickly on bandwagons (FTL neutrinos come to mind), I donate think that the claims here are extraordinary. We used to have a single data point for the formation of planetary systems, so it is not surprising that the observation of other systems has changed what we thought was possible.

Also, there are many different ways in which exoplanet are detected, so it would be surprising that all these methods "conspire" to produce the same kind of incorrect data.
 
  • Like
Likes berkeman, Mlesnita Daniel and jim mcnamara
  • #3
PaulCam said:
Another example is the higher than expected number of very close and very large planets.

Can you cite a prediction for this?
 
  • #4
I think OP is referring to the situation as it was in the 1990s. With just one planetary system as a reference, and way lower computing power, scientists had much simpler planet formation models than nowadays. Inner planets were expected to be small and rocky, outer – massive and gassy. It was no surprise to anyone that massive and close-in planets were first to be discovered by the Doppler method, because this is exactly what you need for strong RV signal. But the first class of discovered exoplanets (actually after pulsar planets) unexpectedly combined these two characteristics. People did not predict migration – Jovian planets were expected to form and stay beyond the snow line. That was the time of wild guesses about exoplanets – some supposed that 51 Peg b and its kind were enormous rocky planets (150 Earth masses!). Now we are in much different situation.

The pool of knowledge gathered over >20 years is enormous. Some hot jupiters are studied to such detail that we observed weather patterns on them. Yes, sometimes scientists jumped to conclusions over weak signals corresponding to interesting orbits and some discoveries were retracted. But almost 4000 are considered confirmed. The papers often start with long discussion about signal analysis, noise reduction, massive unseen companions, starspots and other forms of stellar activity, before claiming discovery of a planet. Thanks to Kepler mission we finally found these small planets that were missing – they were hiding in the noise, and still are easier to spot transiting the stars than to find in RV signal. Large planets on far orbits, on the other hand, are much less likely to transit (it's just geometry) and if observed over a fraction of orbital period, they are hard to confirm.
 
  • Like
Likes Mlesnita Daniel, eloheim, DrClaude and 1 other person
  • #5
PaulCam said:
...

Take for example gas giants that have orbital periods of only a few days. Another example is the higher than expected number of very close and very large planets. The astrophysics community seem to be going all out to find explanations for these extra-ordinary examples...

I believe the rapid orbit planets provide more reliable data. Science is done by reproducing results. If it takes many years for a planet to orbit then it takes many years to repeat the observation. The close in planets are easier to see. The close, large planets have a much stronger gravitational effect on their host star. The hot Jupiters will have the loudest, brightest, and most reproducible signal.

Why pick on exoplanets? The LIGO detectors are looking for stellar mass objects (black holes) that orbit each other hundreds of times per second. There are also stars that orbit each other within days. Algol (beta persei) has been observed for hundreds of years. Beta persei Aa1 and beta persei Aa2 orbit each other in 2.9 days. Both stars have a radius much larger than the Sun. Io orbits Jupiter in 1.8 days. What is wrong with an orbit of a few days?

PaulCam said:
..before you start coming up with fantastical conclusions like a gas giant with an orbital period of a few days... which does not seem likely!...

What would you consider a likely orbital period for a gas giant?

I would look at what type of gas giants are observed. Then use that data to determine what I believe to be most likely.
 
  • #6
Used to be the only practicable way of hunting exo-solar planets was photographic astrometry, so-patiently accumulating star positions for many years, measuring their 'wriggle'. Peter van de Kamp did much, much work thus. Sadly, it came to naught, as his reports of planets around Barnard's Star turned out to be due to a tiny offset in telescope alignment after cleaning etc. Kin to Hubble's mirror mis-calibration. Weep...

Then came the Doppler detectors, and 'Hot Jupiters' whose cyclic tug on their parent stars could be spotted within hours, days then weeks, not decades. Better, this work could be readily duplicated, errors quickly spotted, residuals accumulated.

Once you know such can be found, other methods get deployed. Eclipsing, lensing etc etc.
And now there's a ruddy zoo of discoveries...
Fun thing about eclipsing extra-solar planets is that the orbital plane is tightly constrained so, if you can get Doppler data too, that's seriously synergic for orbital determinations...

One unsettling thought is that for every star with an 'eclipsing' planet, there must be a dozen or more where the orbital plane does not align with us. May be that finding stars / star-systems *without* any planets becomes unusual...
 
  • Like
Likes sophiecentaur, berkeman and Ratman
  • #7
I can add that the sensitivity limits have moved significantly with all the effort put in this topic in last decades. I remember the chart from 2004, showing prospects for planetary detection in the coming years (now in the past). That one: https://commons.wikimedia.org/wiki/File:Extrasolar_Planets_2004-08-31.png. You may note that Space Interferometry Mission was still in development and TESS has not been conceived yet.
 
Last edited:
  • Like
Likes Nik_2213
  • #8
PaulCam said:
Is it not more likely that there are issues with the evidence gathering?

No. The evidence itself, for the transit detection method, is simply a light curve. Or, more specifically, a measurement of the light of a star over many successive images. This is extremely easy to do, and any problems with the sensor or other equipment would be easy to see since it would affect the entire image. Once you take the measurements, you plot the results on a graph and you get the light curve. The dip in the light curve that results from the transit of an exoplanet is different than other solar effects such as the formation of sunspots. They also repeat themselves, and these repeated, identical dips over time are what we use to make absolutely sure we're looking at an exoplanet transit and not something else. There just aren't any other phenomena that could generate these identical repeated dips.

Now, you could claim that maybe there's some other unknown phenomenon that's generating these dips, but the problem is that these dips are exactly what we would expect in an exoplanet transit. So if it looks like a duck, quacks like a duck, and there isn't another animal that better explains it, then it's probably a duck.

The same applies for other detection methods.

PaulCam said:
Is it not scientific that if an experiment results in completely unexpected or extraordinary results that you first question your experimental approach to double and triple check it is not an artefact of your experiemental design...

They are definitely not artifacts of the experiment's setup or design. I have personally taken a light curve of a hot Jupiter using my own telescope and imaging equipment. The dip in the light curve is quite unmistakable. If they were artifacts, I would expect to see them in the light curves of other stars in the same image. But I don't see them.

PaulCam said:
Take for example gas giants that have orbital periods of only a few days. Another example is the higher than expected number of very close and very large planets. The astrophysics community seem to be going all out to find explanations for these extra-ordinary examples.

I'm afraid you have things backwards. Given amplitude of the dip in the light curve, the duration of the dip, and the shape/duration of the entry and exit slopes, the simplest explanation is that a large object, approximately the size of Jupiter, is orbiting very close to its host star. ALL other possibilities are more speculative and more complicated. A hot-Jupiter type planet may seem like an extraordinary and exotic thing, but it is actually fairly mundane in the context of astrophysics. It's just a big planet with a small orbital radius.

Also note that hot-Jupiters are MUCH easier to find than any other exoplanet. Their large size and mass combined with their small orbital radius and quick orbital period causes very large observable effects relative to smaller planets located further from their host stars. There are almost certainly a far greater number of more 'normal' planets than hot-Jupiters.

PaulCam said:
I think the science is too young and the methods used to vague and speculative, open to too much interpretation to be making the kind of claims that are a weekly occurrence in the popular science media.

I don't agree. I can't think of any other interpretation that would reasonably explain the light curves seen during transit events, and I certainly don't see any part of it that is vague and unreasonably speculative. Could be more specific about what you think is vague and speculative? Essentially all of the smaller pieces of the theory/model (measuring the light, orbital mechanics, geometry of the transit, stellar properties, etc) are the result of well understood science and math that is decades if not centuries old.
 
  • Like
Likes stefan r, davenn and Nik_2213
  • #9
Surely a hot Jupiter would have it's atmosphere stripped off it. I mean I know I could calculate the distance from a star for a 5 day orbit (or even faster) but it's certainly less than half of mercury, so the gaseous atmosphere must be exposed to massive temperatures and solar winds.

That's why I felt, intuitively, it sounded wrong.
 
  • #10
A gas giant also has sufficiently high gravity to hold onto its atmosphere.
 
  • #11
Bandersnatch said:
A gas giant also has sufficiently high gravity to hold onto its atmosphere.
Even if it doesn't, there's a loooot of atmosphere to burn off before you make an appreciable dent in the planet.
 
  • #12
Your objection seems to be that everyone is wrong because their results do not agree with your preconceptions. If you have some calculations that shows "a hot Jupiter would have it's atmosphere stripped off it", please show them. But "intuitively, it sounded wrong" just means "I don't like the answer". And unless you are a planetary scientist by profession, where did you develop this intuition?

Furthermore, you do know there are directly imaged exoplanets, don't you?
 
Last edited:
  • #13
Here is what @Vanadium 50 mentioned:

images.jpg


This is from the NASA site -a 4 planet star system.
Go here: https://en.wikipedia.org/wiki/List_of_directly_imaged_exoplanets
The above is one of 7 images taken at the Keck telescope over 7 years. You can see the planets rotate around the star. Slowly. Way cool.
 

Attachments

  • images.jpg
    images.jpg
    1.9 KB · Views: 682
  • #14
PaulCam said:
Surely a hot Jupiter would have it's atmosphere stripped off it. I mean I know I could calculate the distance from a star for a 5 day orbit (or even faster) but it's certainly less than half of mercury, so the gaseous atmosphere must be exposed to massive temperatures and solar winds.

That's why I felt, intuitively, it sounded wrong.
You could rough it out. The Sun has around 13x the surface gravity vs Jupiter. Within √13 solar radii (around 4) that would be a serious problem for Jupiter. Expansion from heating and rotation would increase the distance.
Orbiting right at the sun's surface would be around 2.5 hours period. So a Jupiter with 2.5 day orbit is a bit safer. You can also have Jupiters with 13 times Jupiter's mass. Their radius does not increase much and usually decreases with mass. A cold 15 Jupiter mass object would take gas from the Sun not the other way around.
 
  • #15
Evaporation of hot Jupiter atmospheres has been confirmed with observations of first transiting planet of the kind, HD 209458 b. With a mass slightly higher than Jupiter and orbit eight times tighter than Mercury's around star a bit hottter than Sun, it is surrounded by giant cloud of escaping hydrogen and can be considered quite extreme. However spectacular it may look, the process is not strong enough to strip more than a few percent of its mass throughout the lifetime of the system.
A planet must have even tighter orbit and lower mass (like Saturn or Neptune) to be completely stripped of its atmosphere, if and only if its star fries it with a very high flux of extreme ultraviolet radiation for a prolonged period. There are several papers comparing theoretical models of the phenomenon with the observed population of hot jupiters and hot superEarth's.
 
  • Like
Likes Drakkith
  • #16
Vanadium 50 said:
Your objection seems to be that everyone is wrong because their results do not agree with your preconceptions. If you have some calculations that shows "a hot Jupiter would have it's atmosphere stripped off it", please show them. But "intuitively, it sounded wrong" just means "I don't like the answer". And unless you are a planetary scientist by profession, where did you develop this intuition?

This is fair. I would never claim to be an astro-physicist. Doubt I would ever have come close to being one. I would call myself, in reference to science, "Qualitatively curious". In that beyond about a high school level I don't do or even attempt the mathematics. However in my 44 years of being curious about science in a qualitative way I have found almost all things to be intuitively reasonable. I have even come across many theories in physics where I personally have had similar thoughts before I seen the science, without the maths that turned out to be theories which real physicists have explored with maths and considered plausible.

So coming here and saying I'm suspicious of something is for when the science I see does not slot into my own intuition but in some way seems counter to it as this is quite rare.

It is not me saying "I think it's all wrong and I am right", it is me saying, "Look this stuff sounds fishy and suspect", "Please convince me or show me where my thinking is wrong so that I can believe it."

As someone mentioned above, if your experiment can only detect large, close objects and you find mostly large close objects it at least means your data is biased. So maybe if/when we can start seeing smaller and more distant bodies it will balance itself out and seem more believable.
 
  • #17
Drakkith said:
there's a loooot of atmosphere to burn off

A Loooot? Sounds Canadian.
 
Last edited:
  • #18
PaulCam said:
at least means your data is biased

This is always the case in astronomy. Look up in the sky. Out of all the visible stars, how many are the most common type? Zero. In fact, the closest one would need to be 8 times closer (and thus 64 times as bright) for us to see it.
 
  • #19
Vanadium 50 said:
This is always the case in astronomy. Look up in the sky. Out of all the visible stars, how many are the most common type? Zero. In fact, the closest one would need to be 8 times closer (and thus 64 times as bright) for us to see it.

Let me guess, old, cold, brown dwarfs? Although are red dwarfs not the most common?
 
  • #20
PaulCam said:
It is not me saying "I think it's all wrong and I am right", it is me saying, "Look this stuff sounds fishy and suspect", "Please convince me or show me where my thinking is wrong so that I can believe it."
It may be worth considering that you are asking for the wrong thing. I suggest science has little to do with belief and a great deal to do with acceptance. The evidence, much of which has been referenced by others posters, makes it clear that - for example - hot jupiters do exist. It is therefore reasonable and rational to accept that hot Jupiters are the best explanation for the current evidence. Personal intution and suspicions have a place in science in initiating investigations, but when evidence becomes available then we accept the direction that evidence points.
I think I understand your perspective, it just happens to be a flawed one. Thus, as an equivalent example, I don't believe in the Big Bang, because I find it philosophically unsatisfying, but I do accept Big Bang Theory as overwhelmingly the best explanation for the mountains of evidence in its favour. Review the points made by the other posters and perhaps comment on what you still find unconvincing regarding exoplanets.
 
  • #21
PaulCam said:
As someone mentioned above, if your experiment can only detect large, close objects and you find mostly large close objects it at least means your data is biased. So maybe if/when we can start seeing smaller and more distant bodies it will balance itself out and seem more believable.

We do see smaller and more distant planets. Hundreds if not thousands have been detected.
 
  • #22
Here's a graph showing some of the exoplanet's we've discovered. Notice that a large proportion have orbital periods of less than 10 days, which is to be expected since those are by far the easiest to detect. But also note that roughly half of the exoplanets on this chart have orbital periods of more than 10 days, many having orbital periods of 1000 days and beyond. Also, the overwhelming majority of these exoplanets have masses much smaller than Jupiter.

ExoplanetPopulations-20170616.png
 

Attachments

  • ExoplanetPopulations-20170616.png
    ExoplanetPopulations-20170616.png
    51.1 KB · Views: 567
  • Like
Likes Ophiolite and DrClaude

1. What are exoplanets?

Exoplanets are planets that orbit around stars other than our Sun. They are also known as extrasolar planets.

2. How do scientists detect exoplanets?

Scientists use various methods to detect exoplanets, including the transit method, radial velocity method, and direct imaging method. The transit method involves measuring the dip in a star's brightness as a planet passes in front of it. The radial velocity method measures the slight wobble of a star caused by the gravitational pull of an orbiting planet. Direct imaging involves capturing images of the exoplanet itself.

3. Why is there suspicious evidence surrounding exoplanets?

There have been cases where scientists have reported the discovery of exoplanets based on incomplete or flawed data, leading to skepticism and suspicion surrounding some exoplanet discoveries. Additionally, some scientists have been accused of making fantastical conclusions about exoplanets without sufficient evidence.

4. What are some of the fantastical conclusions that have been made about exoplanets?

Some scientists have claimed that certain exoplanets may have habitable conditions or even support life, based on limited data and assumptions. Others have suggested the existence of advanced alien civilizations on exoplanets, which has been met with skepticism from the scientific community.

5. How can we ensure the accuracy of exoplanet discoveries and conclusions?

To ensure the accuracy of exoplanet discoveries, scientists use rigorous methods and conduct thorough analyses of data. Peer review and replication of results by other scientists also play a crucial role in verifying the validity of exoplanet discoveries. Additionally, it is important for scientists to avoid making sensational claims or jumping to conclusions without sufficient evidence.

Similar threads

Replies
3
Views
2K
Replies
18
Views
3K
  • Computing and Technology
Replies
25
Views
3K
  • Astronomy and Astrophysics
Replies
10
Views
2K
  • Beyond the Standard Models
Replies
4
Views
5K
  • Astronomy and Astrophysics
Replies
6
Views
2K
  • Beyond the Standard Models
Replies
30
Views
7K
  • Astronomy and Astrophysics
Replies
6
Views
14K
Replies
6
Views
3K
  • Set Theory, Logic, Probability, Statistics
Replies
5
Views
2K
Back
Top