How to Find Exoplanets

[steve_bank]

Uhhh,,,Google?


Look for periodic dimming, being done.

Optically resolving a planet has problems with angular resolution and the star light energy washing out the planet's reflected light in the imaging array. I heard something on a show about a technique to compensate for the dynamic range problem.

Spectral analysis to detect H2O absorption of reflected light, according to the show it is being tried.

 

[James Brown]

Aren't we to the point where it's safe to say that any given star is more likely to have planets than not?

 

[repoman]

Well, binary stars may not a high percentage of planets. Same for stars in clusters. Also stars that were ejected from closely packed multiple star systems may not have planets.

But I would guess any star with a history similar to our sun will have many planets.

 

[lpetrich]

Aren't we to the point where it's safe to say that any given star is more likely to have planets than not?

I think so, though the professional literature is rather diffident on that subject. But I've found:

What Fraction of Sun-like Stars Have Planets? - IOPscience

The radial velocities of ~1800 nearby Sun-like stars are currently being monitored by eight high-sensitivity Doppler exoplanet surveys. Approximately 90 of these stars have been found to host exoplanets massive enough to be detectable. Thus, at least ~5% of target stars possess planets. If we limit our analysis to target stars that have been monitored the longest (~15 years), ~11% possess planets. If we limit our analysis to stars monitored the longest and whose low surface activity allows the most precise velocity measurements, ~25% possess planets. By identifying trends of the exoplanet mass and period distributions in a subsample of exoplanets less biased by selection effects and linearly extrapolating these trends into regions of parameter space that have not yet been completely sampled, we find that at least ~9% of Sun-like stars have planets in the mass and orbital period ranges M sin i > 0.3MJup and P < 13 years and at least ~22% have planets in the larger range M sin i > 0.1MJup and P < 60 years. Even this larger area of the log(mass)-log(period) plane is less than 20% of the area occupied by our planetary system, suggesting that this estimate is still a lower limit to the true fraction of Sun-like stars with planets, which may be as large as ~100%.

Prevalence of Earth-size planets orbiting Sun-like stars | PNAS

A major question is whether planets suitable for biochemistry are common or rare in the universe. Small rocky planets with liquid water enjoy key ingredients for biology. We used the National Aeronautics and Space Administration Kepler telescope to survey 42,000 Sun-like stars for periodic dimmings that occur when a planet crosses in front of its host star. We found 603 planets, 10 of which are Earth size and orbit in the habitable zone, where conditions permit surface liquid water. We measured the detectability of these planets by injecting synthetic planet-caused dimmings into Kepler brightness measurements. We find that 22% of Sun-like stars harbor Earth-size planets orbiting in their habitable zones. The nearest such planet may be within 12 light-years.

 

[lpetrich]

Well, binary stars may not a high percentage of planets.

But if the planets are close enough to one of the stars or far enough away from both of them, then their orbits will be stable. That's about 1/3 of their separation for orbiting one planet (S type) and 3 times the separation for orbiting both (P type, circumbinary, "Tatooine"). Nevertheless, in between those stable zones is likely to be a big cleared-out region with only transient objects like comets in it.

Some multiple-star systems are indeed known to have planets, and these planets are present in both kinds of locations.

Same for stars in clusters.

Most star clusters are not dense enough for their stars to pose much of a threat to each others' planets.

One can estimate in a hand-waving sort of way how much density a cluster must have to have before its stars are much of a threat to each others' planets.

One uses a common way of analyzing gases and the like: finding the mean free path of a star between planetary-system-disrupting interactions and then the time needed to travel that path.

(mean free path) = 1 / ( (number density) * (interaction cross section) )

The interaction cross section for disruption I will take to be about 1 AU². The worst case for number density of stars in a cluster is in the center of a globular cluster or a galactic core, and that is 1000 per cubic parsec, or 10^-13 AU^-3. This gives a distance of 10¹³ AU or 5*10⁷ parsecs to travel before disrupting a planetary. Using 100 km/s for a relative velocity (relatively high for solar-neighborhood relative velocities, but low for orbital velocities), I find that it is 100 parsecs per million years or 10⁵ parsecs per billion years. Thus, in the worst case, it would be about 500 billion years before a planetary system gets disrupted by a close encounter of the stellar kind.

 

[steve_bank]

What I said elsewhere.

From spectrum ananlysis and mass estmates stars fall into catgegories.

If stars in similar categories foem by the same processes, planets are a given, at least in our star category.

Proto disks have been obseved.

We have to assume our science applies light years away, indirectly proven by observation and analysis.

As to life in any form, a slightly different matter but similar reasoning.

Abiosphere on Earth took a long time, and an early organism tat helped create O2. A whole lot of probabilities and sernsipiyu involved. Hard to give a probability to Earth like environments.

Organisims exist at black smoker volcanic vents on chemicals deadly to us. Probabiliyu of life in general IMO is high over the scope of the universe.

If evolution and DNA is a constant as with star formation than there is a diversity of life. On the other hand Eath may be one out of an infnite universe.....

From a show on the BB a simulation was run starting at the early particle level, It resulted in objects something resembling galaxies. Takes a lot computer time at the particle level to get any kind of definition. Masybe someday predicting planets and ecosystems.

 

[lpetrich]

What I said elsewhere.

From spectrum ananlysis and mass estmates stars fall into catgegories.

If stars in similar categories foem by the same processes, planets are a given, at least in our star category.

Proto disks have been obseved.

There is a problem. In the early 20th cy., a popular theory for the formation of planets was the tidal or near-collision theory. Two stars would almost collide and they would pull material out of each other. This material would then condense into planets.

This hypothesis made planets *very* improbable. I will now estimate how improbable, using some common gas-physics expressions:
(mean free path) ~ / ( (number density) * (interaction cross section) )

In the local solar neighborhood, the number density of stars is about 0.14 per cubic parsec, and the interaction cross section for creating planets is about pi*(star diameter)^2 or about 10^(-14) square parsecs. This gives a mean square path of 10^(15) parsecs. For a relative velocity of 100 km/s or 100 parsecs per million years, that because 10^(19) years. Dividing by the number of stars in our Galaxy, 10^(11), gives 10^8 years. So somewhere around 100 stars in our Galaxy may have planets.

But the nebular hypothesis has become so strongly established that "hypothesis" does not do its strength justice.

 

[lpetrich]

We have to assume our science applies light years away, indirectly proven by observation and analysis.

More like extrapolation, but it's a very strong inference, at least in the nearer parts of our Galaxy.

Spectral lines are an obvious one -- different laws of physics may produce different spectral lines. The structure and evolution of stars is another. That requires some complicated physics, but it can be tested in various ways. Stars in clusters likely formed at the same time, because there is not much evidence of new stars being formed in most clusters, except for those with lots of new stars forming. One can plot the stars' temperatures and luminosities, and one finds stars moving from the main sequence to red gianthood. One can also do asteroseismology, looking for starquakes. Their periods depend on the stars' internal structure, like their sizes and interior sound velocities, and they are pretty much what one would expect. Not only main-sequence stars fit structure calculations, but also white dwarfs.

 

[lpetrich]

Abiosphere on Earth took a long time, and an early organism tat helped create O2. A whole lot of probabilities and sernsipiyu involved. Hard to give a probability to Earth like environments.

Oxygen release was a latecomer -- it only became geologically noticeable about 2.45 billion years ago, with some minerals becoming more oxidized (The Continuing Puzzle of the Great Oxidation Event - ScienceDirect).

Organisims exist at black smoker volcanic vents on chemicals deadly to us. Probabiliyu of life in general IMO is high over the scope of the universe.

Check out extremophiles some time -- some organisms are adapted to conditions that most others would find intolerable. Like extreme heat or extreme acidity or extreme salinity. But nevertheless, all known Earth organisms require liquid water to metabolize and grow and reproduce, even if they can survive its absence by going into suspended animation.

If evolution and DNA is a constant as with star formation than there is a diversity of life. On the other hand Eath may be one out of an infnite universe.....

Evolution? Yes. It does not depend on its physical substrate; it only needs some form of heredity. DNA? I'm sure that that's specific to Earth biology, and that extraterrestrial organisms will likely have other carriers of hereditary information.

 

[Jokodo]

Well, binary stars may not a high percentage of planets.

But if the planets are close enough to one of the stars or far enough away from both of them, then their orbits will be stable. That's about 1/3 of their separation for orbiting one planet (S type) and 3 times the separation for orbiting both (P type, circumbinary, "Tatooine"). Nevertheless, in between those stable zones is likely to be a big cleared-out region with only transient objects like comets in it.

Some multiple-star systems are indeed known to have planets, and these planets are present in both kinds of locations.

Same for stars in clusters.

Most star clusters are not dense enough for their stars to pose much of a threat to each others' planets.

One can estimate in a hand-waving sort of way how much density a cluster must have to have before its stars are much of a threat to each others' planets.

One uses a common way of analyzing gases and the like: finding the mean free path of a star between planetary-system-disrupting interactions and then the time needed to travel that path.

(mean free path) = 1 / ( (number density) * (interaction cross section) )

The interaction cross section for disruption I will take to be about 1 AU². The worst case for number density of stars in a cluster is in the center of a globular cluster or a galactic core, and that is 1000 per cubic parsec, or 10^-13 AU^-3. This gives a distance of 10¹³ AU or 5*10⁷ parsecs to travel before disrupting a planetary. Using 100 km/s for a relative velocity (relatively high for solar-neighborhood relative velocities, but low for orbital velocities), I find that it is 100 parsecs per million years or 10⁵ parsecs per billion years. Thus, in the worst case, it would be about 500 billion years before a planetary system gets disrupted by a close encounter of the stellar kind.

1 AU² seems low for the cross section?

 

[lpetrich]

1 AU² seems low for the cross section?

A passing star has to get rather close to a planet to strip it away from its star.

Gravitation acceleration of a planet by its star: GM/a² (star mass M, planet distance a)
By a passing star: GM'*a/b³ (star mass M', closest-approach distance b)
Ratio: (M'/M) * (a/b)³

So b must be close to a for disruption.

 

[Bomb#20]

1 AU² seems low for the cross section?

A passing star has to get rather close to a planet to strip it away from its star.

Gravitation acceleration of a planet by its star: GM/a² (star mass M, planet distance a)
By a passing star: GM'*a/b³ (star mass M', closest-approach distance b)
Ratio: (M'/M) * (a/b)³

So b must be close to a for disruption.

In that case, 1 AU² is high for the cross section. Not only do Kepler et al. appear to imply that most stars have planets, but they mostly seem to have planets a lot closer in than Earth.

 

[lpetrich]

1 AU² seems low for the cross section?

A passing star has to get rather close to a planet to strip it away from its star. ....

In that case, 1 AU² is high for the cross section. Not only do Kepler et al. appear to imply that most stars have planets, but they mostly seem to have planets a lot closer in than Earth.

That's to strip away *all* of a star's planets, including the very near ones.

However, the close planets and the massive planets are the easiest ones to detect, and that's why we detect them. The Solar System's planets continue to be only borderline detectable with our technology -- borderline at best.

 

[Jokodo]

1 AU² seems low for the cross section?

A passing star has to get rather close to a planet to strip it away from its star.

Gravitation acceleration of a planet by its star: GM/a² (star mass M, planet distance a)
By a passing star: GM'*a/b³ (star mass M', closest-approach distance b)
Ratio: (M'/M) * (a/b)³

So b must be close to a for disruption.

I know that. I actually wrote, sometime last year, a little suit of Python scripts(1) to simulate what happens if a rogue gas planet / brown dwarf / small red dwarf passes the inner solar system, and saw for myself that nothing much happens most of the time. It actually took quite a number of simulations before even an object of 0.1 solar masses ever managed to kick any one of the terrestrial planets out of the system for good. And I'm aware that I designed my initial conditions to be maximally disruptive: The brown dwarf or star passes with a low relative velocity and almost in the plane of the solar system, giving it the maximum possible time to interact with our planets. I didn't run the model for an encounter with larger stars similar in size to the solar system, but I can see that even then, nothing much would happen much of the time.

But these are the short term consequences. What about the long term consequences? The passing star doesn't have to kick any planet out of the system immediately. If it merely throws Jupiter into an excentric orbit (and if it's massive enough compared to it to tug the sun significantly off course, it can do that without coming close to Jupiter itself), Jupiter will finish its job by destabilising the inner planets' orbit over the course of the next few short million years. If it plows through the asteroid belt, even as the orbits of the planets are barely affected, this might be enough to sterilise Earth...

(1) I did an ugly hack for the initial vectors because I don't understand spherical trigonometry; only because all my planets have low inclinations is the error relatively small. The model for the evolution of the system and interactions between bodies is however sound.

 

[steve_bank]

Oxygen release was a latecomer -- it only became geologically noticeable about 2.45 billion years ago, with some minerals becoming more oxidized (The Continuing Puzzle of the Great Oxidation Event - ScienceDirect).


Check out extremophiles some time -- some organisms are adapted to conditions that most others would find intolerable. Like extreme heat or extreme acidity or extreme salinity. But nevertheless, all known Earth organisms require liquid water to metabolize and grow and reproduce, even if they can survive its absence by going into suspended animation.

If evolution and DNA is a constant as with star formation than there is a diversity of life. On the other hand Eath may be one out of an infnite universe.....

Evolution? Yes. It does not depend on its physical substrate; it only needs some form of heredity. DNA? I'm sure that that's specific to Earth biology, and that extraterrestrial organisms will likely have other carriers of hereditary information.

Yes, evolution is inclusive and we then get into the debate tic define life. DNA-evolution as a constant on planets similar to Earth that may exist.

There was a recent show on Earth's history and geology leading to O2 today. A lot of serendipity and. Odds of another Earth like biosphere much less that
n odds of planets.

 

[lpetrich]

I'll test the hypothesis of a "Kepler dichotomy" between 1-planet and multiplanet systems, or more precisely, systems where we observe only one planet and where we observe more than one planet.

The observed number of stars with each number of planets: {1204, 286, 103, 39, 15, 3, 0, 1}

I did a chi-squared fit, setting the variance for 7 planets to 1 -- it is normally the number to be fit to.

For all the data values: 17.8954 with df = 6 -- 0.0065 chance of being greater
For all but the first one: 3.9464 with df = 5 -- 0.56
For all but the first two: 3.43395 with df = 4 -- 0.49
df = degrees of freedom = # data points - # parameters to be fit

So an exponential distribution gives a good fit if one omits the count of 1-planet systems. That fit gives a 1-planet estimate of 836 systems, giving a discrepancy of 368 systems, about 44% of the estimate.

 

[lpetrich]

TESS is on its way: NASA Planet Hunter on Its Way to Orbit | NASA

Its launch was delayed two days because of the weather, but the weather was very nice today at Cape Canaveral. I'm old enough to remember when it was renamed Cape Kennedy.

It will spend the next two months getting into its planned orbit, a very elliptical one that extends out to the Moon's distance and that will have half the Moon's orbit period. The spacecraft will reach maximum distance when the Moon is about 90d ahead or behind, making that orbit stabilized by the Moon's pull on the spacecraft. It will do its observations in the outer part of its orbit and upload its results in the inner part of its orbit.

It will observe a strip from low southern ecliptic latitude to the south ecliptic pole, and for the next 12 months, it will then be turned to observe a neighboring strip. After its first year, it will look at the northern ecliptic hemisphere for another year of observation. It may then continue to do even more searching. Ecliptic = plane of the Earth's orbit around the Sun.

Though a month is not much time coverage, it should suffice for planets of red dwarfs. At high ecliptic latitudes, the observation strips should overlaps, and stars at the ecliptic poles should be observed almost continuously for a year. Kepler's observation spot is near the north ecliptic pole, so TESS should be able to observe some of the Kepler stars.

TESS will be looking for planets around some of the brighter stars, because it is easier to get the stars' spectra that way. That will be done for radial-velocity measurements, to get the planets' masses, and also for observing transits in several wavelength bands, for getting an idea of what the planets' atmospheres are like, if they have atmospheres.

 

[lpetrich]

TESS Continues Testing Prior to 1st Observations | NASA

The TESS team has reported that the spacecraft and cameras are in good health, and the spacecraft has successfully reached its final science orbit. The team continues to conduct tests in order to optimize spacecraft performance with a goal of beginning science at the end of July.

NASA’s Kepler Spacecraft Pauses Science Observations | NASA

Earlier this week, NASA’s Kepler team received an indication that the spacecraft fuel tank is running very low. NASA has placed the spacecraft in a hibernation-like state in preparation to download the science data collected in its latest observation campaign. Once the data has been downloaded, the expectation is to start observations for the next campaign with any remaining fuel.

The spacecraft was in its 18th observing run of its K2 extended mission.

To bring the data home, the spacecraft must point its large antenna back to Earth and transmit the data during its allotted Deep Space Network time, which is scheduled in early August. Until then, the spacecraft will remain stable and parked in a no-fuel-use safe mode. On August 2, the team will command the spacecraft to awaken from its no-fuel-use state and maneuver the spacecraft to the correct orientation and downlink the data. If the maneuver and download are successful, the team will begin its 19th observation campaign on August 6 with the remaining fuel.

A changing of the guard.

 

[lpetrich]

Exoplanet-Hunting TESS Starts Science Operations | NASA

NASA’s Transiting Exoplanet Survey Satellite has started its search for planets around nearby stars, officially beginning science operations on July 25, 2018. TESS is expected to transmit its first series of science data back to Earth in August, and thereafter periodically every 13.5 days, once per orbit, as the spacecraft makes it closest approach to Earth. The TESS Science Team will begin searching the data for new planets immediately after the first series arrives.