How to detect a new world

This is a bilingual post. For a Portuguese version click here.

One. Zero. Zero. Zero. Zero. Zero. Zero. Zero. Zero. Zero. Zero. Zero. That’s 100,000,000,000, or 100 billion. That is the amount of stars we believe orbit our galaxy, and yet, we know of 2609 that are orbit by exoplanets. This is comparable to entering a huge room with the whole population of the states of California, Texas, Florida and Illinois together, and only being able to see two and a half people. Here is why.


Exoplanets or extra solar planets are the names given to celestial bodies that orbit stars other than our Sun, but detecting them is not an easy task and frequently we need to count with different indirect methods to know they’re there.

In 1990, using calculations to measure pulsar radio emissions, scientists detected a variation on the signal of PSR B1257+12, a pulsar 2.3 thousand light-years from Earth. A pulsar is a very unique celestial body: their signals are supposed to be rock steady, and variations need an explanation. Once enough data was accumulated, using Arecibo’s radio-telescope in New Mexico, the scientists responsible for the analysis were capable to work out that the only explanation for this pulsar’s signal variation was two big planetary masses. This was the first method capable to confirm a planet orbiting another body – even if not a star. But pulsars are somewhat rare in our galactic neighbourhood, and soon looking for pulsars’ wobbles was not the only way to detect exoplanets. New methods were needed to continue and improve the search for exoplanets. The method of radial velocity was one of them.

Even though a few detection methods existed on the early 2000’s – we could confirm three exoplanets through direct imaging still on 2004 – radial velocity was our favourite method to find planets around other stars since 1995, because it was the most reliable known method at the time, and still is the most used method when searching from ground telescopes.

Radial velocity measures the wobble of the star by analyzing it’s light spectrum. A shift to red identifies that it is moving away from us, while a shift to blue indicates that the star is moving towards us. Since this shift is caused by the Doppler effect, this method is also known as the Doppler Method. How much the star light shifts to red or blue, determines the amount of mass or masses that orbits it. According to data from NASA’s exoplanet’s archive, until 2008 radial velocity alone was responsible for almost 80% of the 326 confirmed exoplanets. Another method that gained a lot of popularity after 2004 was transit photometry, back then responsible for around 40 confirmations per year. Then, in 2009, the spacial observatory Kepler goes to orbit and changes the history of exoplanets detection.

Kepler uses transit photometry to detect exoplanets, but because it is in orbit the observatory doesn’t suffer the same issues with light pollution that the ground observatories do. On transit the luminosity of a star is surveyed. If the star experiences a certain drop on its luminosity, it then goes to a list of potential solar systems for later confirmation. The bigger the drop on luminosity, the larger is the exoplanet. We can think of a transit like a micro-eclipse, but from such a distance and so faint that only a space observatory with electronic eyes can see. After the first detection the star is surveyed until at least three of those transits are measured, and then the star is confirmed as being a solar system.

But those confirmations can take a long time. An exoplanet with orbit similar to our Saturn would take 29 years to complete its orbit, which means that taking in account our current technology and methods, would take 87 years to confirm that it exists! Today with Kepler’s help, 80% of the exoplanets are detected via transit. While this post is written we can count 3499 exoplanets confirmed since 1992 through different methods, orbiting 2609 stars.

Amount of detected exoplanets, by method, per year

If this number seems too low when compared to the approximately 100 billion stars on our Milky Way is because it is. Why can’t we detect more?

Plain and simple, because the detection methods are limited by the same physical rules that we are, no matter where on the Universe we are searching.

Both the transit photometry and radial velocity are limited by the orbital plane angle of the stars being observed. Orbital plane is an imaginary disk around the star, where most or all planets are expected to orbit, like this:

Our Solar System’s orbital plane. Image not to scale.

To observe the light spectrum shift, it has to be moving away or towards us when it’s ‘pulled’ by the mass of the exoplanet. If we are looking from “the top” of the orbital plane, the star is not moving towards or away, it’s moving in a circle from our perspective:
Screen Shot 2017-08-09 at 3.54.36 PM

Since radial velocity relies on detecting this shift on the star’s light spectrum, all detections done via radial velocity are of stars with an orbital planes at least at a small angle to us, never from ‘the top’.

This is even more evident on transit. As we mentioned, this is comparable to a micro-eclipse from far away, and therefore it can only be perceived when the exoplanet is crossing the front of the star to our perspective. A very, VERY narrow source of light:

Screen Shot 2017-07-30 at 9.59.33 PM
The exoplanet’s orbit must be almost edge on to allow detection!

As one can imagine, there is no reasons for orbital planes to be edge on to us. The methods behind 96% of the combined confirmed exoplanets can only detect stars with orbital planes at a certain position in relation to us. Is like entering a room full of people, but only being able to see the ones with their shoulders pointing to you. Add to this the size of Kepler’s field of view:

Kepler’s field of view

And not on only seeing people with their sides to us, we are looking at the room with a mask covering our eyes, with a tiny hole to look through. Finally, add the distance Kepler can survey stars:


Here we have the final comparison: looking into a huge room the size of a baseball field, but we can see as far as the catcher’s box, through a small hole on a black mask and for two seconds. The field is crowded with 100 million people, from which we can see only  two and a half.


This is why there are so few detected extra solar planets known. Plus, we have been doing that for around 20 years, and actually well for 10 or so. If we were alpha-centaurians, 4.6 light-years from here, we couldn’t have confirmed Jupiter yet since it have a 12 years orbit. But of course this doesn’t end here: NASA expects to put TESS (Transiting Exoplanet Survey Satellite) in orbit in 2018. It will survey all the sky for stars and provide the list of potential stars for  Kepler and other observatories for a deeper analysis. And there’s James Webb Space Telescope also scheduled to launch in 2018. JWST is the larger space telescope ever to be put in space, with a 21 feet mirror, vs Hubble’s 8 feet mirror.

In a certain way we are blind to the majority of the galactic neighbourhood, but our technology doesn’t stop improving. Perhaps a new method is waiting to be discovered or enhanced, and changes again the race for exoplanets.

For now, we will keep searching.

Per ardua ad astra (Through struggle to the stars)


Wolszczan, A.; Frail, D. (1992). “A planetary system around the millisecond pulsar PSR1257 + 12”. Nature. 355 (6356): 145–147. Bibcode:1992Natur.355..145W. doi:10.1038/355145a0.

edX course: Arizona State University: AST111 Introduction to Solar Systems Astronomy


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s