• Welcome to the new Internet Infidels Discussion Board, formerly Talk Freethought.

How to Find Exoplanets

lpetrich

lpetrich

Contributor
Joined
Jul 27, 2000
Messages
26,334
Location
Eugene, OR
Gender
Male
Basic Beliefs
Atheist
It was something that I never expected to happen. All these discoveries of exoplanets happening when they did. NASA Exoplanet Archive lists 3708 confirmed planets, as I write this, with more to come to be found by upcoming searches.

 Methods of detecting exoplanets lists numerous methods, and I will create a family tree of them, sorting them out by method.

  • Direct
  • Indirect
    • Transits
    • Gravity
Direct detection:
  • Imaging in visible / infrared (planet separate from star)
  • Planet and star together
    • Reflection/emission modulation: planets' reflected or emitted light
    • Polarimetry: planets' light being polarized by reflection
    • Modified interferometry to look for a planet's spectrum
  • Radio emissions
    • Magnetospheres
    • Auroras
Transits:
  • Photometry
  • Interferometry to make an image of a star with a transiting planet
Gravity:
  • Line-of-sight motion
    • Pulsar timing
    • Variable-star timing
    • Eclipsing-binary timing
    • Radial velocity
    • Relativistic beaming: focusing of light in the direction of motion
  • Transit timing: from perturbations by other planets
  • Transit duration variation: like transit timing
  • Astrometry: sideways deflection
  • Gravitational microlensing: planets, like stars, can focus the light of stars behind them
  • Ellipsoidal variations: American-football-shaped tidal effects on the planet's star

When I was growing up, I remember Peter van de Kamp's claim that Barnard's Star had one or two planets that he had detected by astrometry with his telescope at Swarthmore University near Philadelphia, PA. But another astronomer failed to detect it, and the largest observed effects turned out to be from maintenance on his telescope. So the first generally-accepted detection of an exoplanet was in 1992 for two planets orbiting a pulsar and in 1995 for a planet orbiting a sunlike star. Carl Sagan barely got to see this new age of astronomy before he died.
 
 List of exoplanet search projects lists many of the better-known ones, both ground-based and space-based. Ground-based searches are much cheaper than space-based ones, and equipment maintenance is much easier, but the Earth gets in the way much of the time, and its atmosphere can have clouds in it. That explains why Chile has become popular for doing astronomy -- its Atacama Desert is *very* dry.

The best-known exoplanet-search spacecraft is the Kepler one, and its observation of some 190,000 stars has yielded the discovery of over 2000 planets. Although that is 1 planet per 100 stars, that is because transits depends on planets' orbits being nearly edge-on relative to us.

The Kepler spacecraft is currently in its K2 mission, the successor of its main mission, but the spacecraft is running low on reaction gas, and its mission will come to an end in a few months or so.

But there are others on the way.

TESS - Transiting Exoplanet Survey Satellite is due to be launched in mid-April, and it will look for planet transits of the brighter stars of all over the sky.

CHEOPS Mission Homepage (ESA Science & Technology: CHEOPS: CHaracterising ExOPlanets Satellite) is a mission for doing precision observations of transits of some known exoplanets, for the purpose of better understanding planets with sizes between Earth and Neptune sizes (1 - 4 Earth sizes). Super-Earths and mini-Neptunes are one of the surprising discoveries of the Kepler mission; there are no Solar-System planets that fall into that range.

The James Webb Space Telescope is to do observations of exoplanet transits in several wavelengths, for getting a clue as to their atmospheres' compositions. It was recently scheduled to launch in 2019, and it is now scheduled for May 2020.

Wide Field Infrared Survey Telescope (WFIRST) will do direct imaging and search for gravitational-microlensing events.

ESA Science & Technology: PLATO (PLAnetary Transits and Oscillations of stars) is another transit searcher.

UK will lead European exoplanet mission - BBC News notes ESA Science & Technology: ARIEL (Atmospheric Remote-sensing Infrared Exoplanet Large-survey) will look at transits over several wavelengths, as JWST will do.
 
All of them.

I do wonder, though- what's the most distant planet we've definitely spotted? I rather suspect it's one of the 'giant hot Jupiter' varieties, orbiting close to its parent star.

I well recall when fp, the term in the Drake equation for the number of stars with planets, was entirely unknown. It startles and delights me that in the past couple of decades we've discovered that planets are amazingly plentiful!
 
Using the info at the Exoplanet Archive lpetrich links to, I find that the two most distant planets are SWEEPS 4 b and SWEEPS 11 b- both of which are 8500 parsecs (almost 28,000 light years) away! And my guess was right- 4 b is listed as having a period of 4.2 days, and a mass 3.8 times that of Jupiter; and 11 b is nearly ten times as heavy as Jupiter (with considerable uncertainty) and orbits its star in a mere 1.796 days!

(Cross post with fast. :))
 
I wonder what two planets are fartherest from each other. We can't tell with distance alone. For example, if there is a car 7 miles from me and one 3 miles from me, they wouldn't be 4 miles apart if the car 7 miles away is to my east and the car 3 miles away is to my west.
 
 Sagittarius Window Eclipsing Extrasolar Planet Search (SWEEPS) was a search for exoplanet transits done in 2006 with the Hubble Space Telescope. That search was done in the galactic bulge, using a line of sight with relatively little interstellar dust. Some 16 planets have been found, all Jupiter-sized, planets like SWEEPS-11.

The SWEEPS planets are the farthest ones considered well-established, but  Extragalactic planet notes some gravitational microlensing events that point to planets outside our Galaxy.
 
What is the orientation of Barnard's Star rotational plane to our solar rotational plane?

If edge on then transit or wobble method would theoretically work?

If head on then existing planets would be much harder to find, right?

https://arxiv.org/pdf/astro-ph/9806276.pdf

looks like it would not be easy or even possible to have it with current data.
 
fast said:
I wonder what two planets are fartherest from each other.
Click to expand...
Planets orbiting the same star or planets orbiting different stars? I'm confused.


The known planet that is the farthest from its star is WD 0806-661 b.
  • Projected separation: 2500 AU
  • Planet mass: 7.5 +- 1.5 MJup
  • Planet temperature: 322.3 +- 22.3 K
  • Star mass: 0.62 MSun
  • My orbit-period estimate: 160,000 years
he star's mass, about 0.62 solar masses, but about nothing else. That gives an estimated period of about 160,000 years.

The second-farthest one, GU Piscium (Psc) b, has more info available on it and its star.
  • Projected separation: 2000 +- 200 AU
  • Planet mass: 11.3 +- 1.7 MJup
  • Planet radius: 1.265 +- 0.115 RJup
  • Planet temperature: 1050 +- 50 K
  • Star mass: 0.325 +- 0.025 MSun
  • Star spectral type: M3
  • Star age: 0.100 +- 0.030 Gyr
  • My orbit-period estimate: 160,000 years
 
Different stars.

If I said which two Walmarts are furtherest apart, it wouldn't do me any good to look at the two most fartherest from me.
 
fast said:
Different stars.
Click to expand...

To do a good job of this, I'd have to download a data archive, and then crunch its numbers. fast, you ought to try that yourself, if you can.

OK. Aside from extragalactic microlensing, the farthest-known exoplanets are the SWEEPS one. I'll use SWEEPS-11, which is at
RA = 17h59m02.00s
Dec = -29d11m53.5s

Its antipode is at
RA = 5h59m02.00s
Dec = 29d11m53.5s

My best guess is CoRoT-20 b:

RA = 06h30m52.90s
Dec = +00d13m36.9s
Gal Lng = 210.38124d
Gal Lat = - 4.46853d
Distance = 1230+-120 parsecs

The planet's star is a very Sunlike one, and the planet itself is a "hot Jupiter", orbiting at around 0.09 AU. Its equilibrium temperature is around 1000 K -- hot enough to glow in visible right.
 
I expect there's 3-D maps showing all the known planets, and you could easily enough spot which two are the furthest apart just by looking at one.
 
Thinking about SWEEPS 11 b- if it takes less than two days to orbit its sun, I would say it must be within the solar corona, and thus quite rapidly being slowed (on a cosmological scale) and dragged ever closer to a collision with its primary. Given the numbers of hot Jupiters we've found, I suppose such collisions must be fairly common, galaxy-wide. How would we be able to distinguish such an event from, say, an ordinary nova?
 
 SWEEPS-11, The Extrasolar Planet Encyclopaedia — SWEEPS-11:

a = 0.03 AU = 6.4 RSun
Period = 1.796 days
Planet mass = 9.7 +- 5.6 MJup
Planet radius = 1.13 +- 0.021 RJup
Star mass = 1.1 MSun
Star radius = 1.45 RSun

For the corona density, I used the Sun's value in The Sun and its Corona - AndreHandout.pdf for an altitude of about 6.4 million km. That gives a number density of about 10^6 hydrogen atoms per cubic centimeter, and a mass density of about 10^(-18) g/cm^3.

The planet's orbital velocity is about 180 km/s, and I calculate that the planet sweeps up about 10^(-14) of its mass each year. For the planet to fall in over a billion years, one needs 10^6 times more density, or 10^12 hydrogen atoms / cm^3. For the Sun, that is reached for the chromosphere-corona transition region, about 2000 km up.

But a likely fate is for the planet's volatile contents to be stripped off first, leaving behind the rock and iron in its core. Some very close rocky planets may be such "chthonian planets".
 
Jobar said:
I well recall when fp, the term in the Drake equation for the number of stars with planets, was entirely unknown. It startles and delights me that in the past couple of decades we've discovered that planets are amazingly plentiful!
Click to expand...

If you want to observe evidence of exoplanets, you can do that with a telescope with a digital camera: Exoplanet Section | aavso.org -- that's the American Association of Variable Star Observers. You'll mostly observe transits of hot Jovian planets, but you'll get to observe these oddities for yourself.
 
My favorite technique for discovering exoplanets is Transit Timing Variations, TTV's. Discovering planets by their perturbations of the orbits of known ones.

In the Solar System, Urbain Leverrier and John Couch Adams successfully predicted the presence of Neptune from discrepancies in the orbit of Uranus.

Their successors went on to predict the existence of an additional planet, and Pluto seemed to be that planet at first. But it was discovered to be too small, and when Voyager 2 flew by Neptune, tracking that spacecraft enabled improving the planet's mass value, and that improvement made the discrepancy go away.

Leverrier also predicted the presence of an intra-Mercurian planet or planets from Mercury's excess orbit precession. But no intra-Mercurian planets or asteroids were ever found, and some astronomers proposed modifications of the law of gravity. Einstein's General Relativity successfully explains it -- and it is a modified-gravity theory.

Turning from planets to moons, moon-moon perturbations have been used to find the masses of the larger moons of Jupiter and Saturn. This is helped by the orbit resonances that some moons are in. Io, Europa, and Ganymede are in a 1:2:4 resonance, Mimas and Tethys in a 1:2 resonance, Enceladus and Dione also in a 1:2 resonance, and Titan and Hyperion in a 3:4 resonance. Resonances amplify the effects of their interactions, though this amplification is over several orbits.


In 2011, astronomer Sarah Ballard joined Urbain Leverrier and John Couch Adams in predicting a planet. She searched for planets in the abundant data returned by the Kepler space telescope, and she found four of them, three of them from transits: Kepler-19 b, Kepler-61 b, and Kepler-93 b. But there was something odd about Kepler-19 b. It was sometimes fast in its orbit, sometimes slow. After considering various possibilities, she concluded that there is an additional planet in the system, one in an orbital resonance with the already-observed planet. This planet is now called Kepler-19 c, and it is her fourth discovered planet. Her discovery paper: [1109.1561] The Kepler-19 System: A Transiting 2.2 R_Earth Planet and a Second Planet Detected via Transit Timing Variations A later paper's authors use radial-velocity data and find an additional planet: [1703.06885] The Kepler-19 system: a thick-envelope super-Earth with two Neptune-mass companions characterized using Radial Velocities and Transit Timing Variations

Since then, several other planets have been discovered using TTV's, and TTV's have been used to estimate the masses of several known ones, like the seven TRAPPIST-1 planets.
 
If "only" 1% of the stars we see have an edge-on planetary orbital path and there are another 180 degrees of unseen paths, doesn't that statistically mean that pretty much every star has planets orbiting it?
 
Malintent said:
If "only" 1% of the stars we see have an edge-on planetary orbital path and there are another 180 degrees of unseen paths, doesn't that statistically mean that pretty much every star has planets orbiting it?
Click to expand...
Pretty much, or at least a large fraction of them. But it's been hard for me to find good numbers for that.

The Kepler space telescope observed some 160,000 stars in its primary mission, and the analyzers of its data have found 2342 planets that they consider confirmed. I will now count how many stars that the planets orbit. I'll use statistics on multiplicity, again from NASA's exoplanet archive.

Kepler's main mission: 1: 1204, 2: 286, 3: 103, 4: 39, 5: 15, 6: 3, 7: 0, 8: 1 -- 2342 planets around 1651 stars
Kepler's extended mission (K2): 1: 174, 2: 31, 3: 14, 4: 3, 5: 2, 6: 0, 7: 1 -- 307 planets around 225 stars
Non-Kepler transits: 1: 288, 2: 1 -- 290 planets around 289 stars
Radial velocity:1: 396, 2: 72, 3: 16, 4: 8, 5: 3, 6: 3 -- 653 planets around 498 stars
Direct imaging: 1: 38, 2: 1, 3: 0, 4: 1 -- 44 planets orbiting 40 stars
Microlensing: 1: 52, 2: 2 -- 56 planets orbiting 54 stars
Pulsar timing: 1: 3, 2: 0, 3: 1 -- 6 planets orbiting 4 pulsars

So out of the 160,000-odd stars that the Kepler telescope observed over its main mission, about 1% were discovered to have planets.
 
You must log in or register to reply here.
Back
Top Bottom