Date: May 25, 2012

Title: Encore: Detecting Moons Around Extrasolar Planets

Podcaster: Adam Fuller

This podcast originally aired on March 22, 2010

Description: As more exoplanets are discovered transiting their star, scientists have begun looking for potential moons orbiting these exoplanets. Today’s podcast is about how we can detect an exomoon and what type of exomoons would be easiest to detect.

Bio: Adam Fuller is currently a graduate student in the Planetary Sciences department at Johns Hopkins University in Baltimore, Maryland. He graduated from Columbia University in the City of New York with a B.S. in Astrophysics in 2009. He also has a B.A. in Journalism from North Carolina. His research interests include planetary science, meteorology, and astrobiology. His favorite planet is Saturn, and he has not seen Avatar. Outside of school, he is an avid marathoner and a dedicated uncle to three proto-astrophysicists.

Today’s Sponsor: “This episode of 365 days of Astronomy” is sponsored by – Expanding your horizons in astronomy today. The premier on-demand telescope network, at dark sky sites in Spain, New Mexico and Siding Spring, Australia.


Hello everyone. My name is Adam Fuller, and I’m a grad student in the Planetary Sciences department at Johns Hopkins University in Baltimore, Maryland. In today’s 365 Days of Astronomy podcast I’ll be discussing how we may someday soon begin detecting moons around extrasolar planets, otherwise known as exomoons.
There are several methods used to detect exoplanets. Some are more successful than others and even fewer are suited for detecting exomoons.

Our most promising technique for detecting exomoons is called the transit method. This is when we observe a planet passing across the face of its star relative to us. Roughly 70 planets have been found this way. It’s akin to an eclipse. When the moon eclipses the sun, whether it’s a partial or total eclipse, the moon moves between the Earth and the Sun and blocks a portion of the light coming from the Sun. We see the drop in light and, using trigonometry, we can determine things like the size and distance of the moon, the sun, and even the period of the moon’s orbit. The same thing applies to detecting planet transits of other stars. We can detect the miniscule drop in light we receive from the star as the planet crosses the disk of the star and blocks some of its light. From this tiny dip in the star’s luminosity, often less than 1% of its total brightness, we can determine a lot of things about the planet. The depth of the dip tells us the radius of the planet. How often the dip occurs tells us the orbital period of the planet. The planet’s semimajor axis, orbital eccentricity, the inclination of its orbital plane, and its mass and density can be inferred from the dip. Variations in when the transit starts and stops can even tell us about the possible presence of other bodies, even non-transiting ones, orbiting that star. And, as current research shows, with sensitive enough equipment we can even detect any potential exomoons and determine their characteristics as well.

How can we do that? When a planet and moon transit across the star and cause a dip in the luminosity, the variations in the depth and length of the dip are a big giveaway. Imagine we’re measuring the luminosity of a star from Earth over several nights. When we plot the luminosity against time, we call this the light curve.

If there are no transiting planets, the plot should roughly be a straight horizontal line. If there is a transit event in our plot, however, it will look like somebody erased part of our line and drew a letter U hanging down from the line. The depth of the U depends on the size of the planet that crosses the star. Just like a solar eclipse with our moon and Sun, from our point of view, it looks like two circles overlap each other. The midtransit time is the middle of the transit, when the planet has passed as close to the star’s center as it ever will during its transit. This is represented as the bottom-most point in our dip. The area of the circle made by the planet is pi times the squared radius of the planet. Divide this area by the area of the circle made by the star, and this will give you the percentage of light blocked by the planet. This tells us how deep the dip in our light curve goes.

If we watched Jupiter transit the sun from outside Jupiter’s orbit, the dip in the light curve would only go down 1.057%. This is detectable, and, in fact, so far the tiniest planet seen transiting its star, CoRoT-7b, is only 15% the size of Jupiter. That’s just 1.7 times the size of Earth. If we could see Jupiter’s four largest moons during this transit—if they weren’t exactly between the Sun and Jupiter or Jupiter and us—they could cause the dip to go down another 0.004%. That’s hardly anything. What about a planet twice the size of Jupiter with an Earth-sized moon orbiting a sun-like star? The planet would cause a 4.226% dip in the light curve, and the Earth-sized moon would cause another 0.008% dip. That’s still not a lot for the moon’s detection, but we can begin to see that, after playing with the numbers, we may have more success looking for moons around stars smaller than our Sun, K and M class stars. Fortunately there’s a lot of those in our galaxy.

The depth of the dip isn’t the only thing we can use from the light curve. Consider the center of mass between the moon and the planet. Our moon doesn’t orbit the center of the Earth. Instead, the Earth and moon both orbit a point only about a quarter of the way under the Earth’s surface towards its core. If you watched the moon orbit the Earth along the moon’s orbital plane from space, you would see the moon moving left to right and back as it passes first in front of and then behind the Earth. Even though it wouldn’t be much, the Earth would move, too. Another example is Pluto and its moon Charon. The center of mass between those two isn’t even inside Pluto’s radius! Watching those two orbit their center of mass would be like watching a dumbbell spin on the floor while laying down next to it. Pass them in front of a star and sometimes you’ll see Pluto start to transit first, other times Charon will transit first. During the transit, Pluto and Charon may eclipse each other so that, from beyond their orbit, it appears as though only Pluto is transiting. This means that each dip in our observed light curve may have it’s own dips depending on where the moon and planet are in their orbits around their center of mass. As a transit begins, the first part of the dip might be very steep as the planet crosses the star first. Or it could be a small shallow dip as the moon swings out in front first. If the transit is long enough relative to the period of the moon’s orbit, then the moon may eclipse its planet during the transit, causing a miniscule increase in luminosity, before exiting the eclipse and reappearing on the face of the star. From this we know that to catch a moon transiting a star, the closer its size is to its planet’s, the easier it is to see its effect on the dip in the light curve. An Earth-sized moon orbiting a Saturn-like planet will be easier to detect than an Earth-sized moon orbiting a Jupiter-like planet. Saturn is 85% the size of Jupiter, but it’s only half as dense. An Earth-sized moon would have more of a gravitational effect on Saturn than Jupiter, so the dip in the light curve will be almost as deep but with more tell-tale fluctuations.

The tiny changes in the light curve due to a moon, however, may be very hard to see with just one transit event. If they’re smaller than the observational error, we would have no idea the moon’s even there. But take several light curves together, and over time we might be able to pull out a periodic signal in how the errors behave. From this signal we can theoretically determine the moon’s mass and its distance from its planet. The trick, however, is to couple the transit data with other data, like radial velocity measurements, to make sure that we’ve actually discovered an exomoon and not rings or even trojan-like asteroids leading or following the exoplanet in its orbit.

To recap, we’ve seen that it’s easier to detect exomoons around planets that orbit small stars, and that the closer the moon and its planet are in mass, the easier it is to detect them doing their dance during a transit. So what are the chances of actually discovering an exomoon? What about the Kepler mission? Recent research says that if the exomoon is at least one fifth of Earth’s mass, with enough time Kepler should see it. Our moon is one sixth of Earth’s mass, so it may not be detectable. In fact, even though Kepler has the ability to detect Earth-sized exoplanets, if we were looking for a Moon-sized object orbiting an Earth-sized planet around a Sun-like star, the transit timing variations in the light curves might still be too small to pop out of the observational error. Earth-sized moons around Jupiter-sized planets? Kepler can detect that.

And research has shown that the exomoon’s orbital period and its planet’s distance from the star may be just as important in detecting the exomoon as its size relative to its planet is. Exomoon orbital periods between 10 and 35 days around Jupiter-mass planets between 0.6 and 1 AU from their star should be relatively easy to detect.

Well, that’s all I have for today. I hope you’ve enjoyed today’s podcast about detecting extrasolar moons. I always welcome questions about this, so the perplexed or curious can email me at afuller at eps dot jhu dot edu. Have a great day and keep listening.

End of podcast:

365 Days of Astronomy
The 365 Days of Astronomy Podcast is produced by the New Media Working Group of the International Year of Astronomy 2009. Audio post-production by Preston Gibson. Bandwidth donated by and wizzard media. Web design by Clockwork Active Media Systems. You may reproduce and distribute this audio for non-commercial purposes. Please consider supporting the podcast with a few dollars (or Euros!). Visit us on the web at or email us at Until tomorrow…goodbye.