Dusting for Prints: New Clues for Exoplanets in Old Hubble Images

Disks of debris detected around young stars using the new technique. The top row represents the images themselves. The bottom row is an artist’s illustration of how those disks might appear, oriented against a background starfield. Image Credit: ASA/ESA, R. Soummer, Ann Feild (STScI)

Disks of debris detected around young stars using the new technique. The top row represents the images themselves. The bottom row is an artist’s illustration of how those disks might appear, oriented against a background starfield. Image Credit: ASA/ESA, R. Soummer, Ann Feild (STScI)

Sometimes when you look at a photograph for a second time you see something for the first time. This is exactly what happened when Dr. Rémi Soummer re-examined Hubble images of 400 stars. In what had previously been empty space, Dr. Soummer and his team found evidence of five young solar systems that are similar to our solar system when the Earth was forming.

“These findings increase the number of debris disks seen in scattered light from 18 to 23. By significantly adding to the known population and by showing the variety of shapes in these new disks, Hubble can help astronomers learn more about how planetary systems form and evolve,” said Soummer.

This find is the result of stellar detective work and gives us more hope than ever of finding habitable, rocky planets around stars in our galaxy.

Planets form from enormous clouds of gas and dust that collapse. The disks formed by the collapse produce stars at the center and give rise to planets at the periphery.

While the planets themselves are not directly seen, the gravitational tug of planets on the dust in the disks acts as a clue to their presence.

“Much like the rings of Saturn are influenced by its moons, planets influence the material in a disk around a star,” said Dr. Dean Hines of the Space Telescope Science Institute, a member of the team that made the discovery, “We look for such features by eye, or by subtracting a smooth ‘model’ and looking at the non-smooth structures that are left over. We can also measure the surface brightness of the disk as a function of distance from the star, to look for gaps and changes in the brightness that might indicate changes in the amount of dusty material.”

Simulations show that planets sweep through stellar disks gathering up dust, pushing it to one side or the other. This can cause the disks to assume odd shapes. A star with an eccentric-looking disk is therefore a good candidate for finding a planet.

The light that NICMOS sees. Image Credit: The Hubble Space Institute, NASA and STScI

The light that NICMOS sees. Image Credit: The Hubble Space Institute, NASA and STScI

In and of itself, the endurance of a stellar dust disk can mean that a planet is on the rise within it. Powerful solar winds should cause the disk of dust to disperse quickly-within less than ten million years. The persistence of a cloud or disk of dust can be accounted for by small proto-planets called planetesimals, which are crashing into one another. This process forms new larger planets, but also replenishes the disk with debris from the collisions. Small bodies such as asteroids and comets left over from the process of terrestrial planet formation can also continue to collide, increasing the longevity of stellar dust disks.

Not only does a lingering disk mean that a planet is more likely to be found, but it also may mean that the planet in question is especially interesting from the standpoint of astrobiology. Certain theories about terrestrial planet formation hold that a persistent disk may mean a calmer stellar environment-one that is more conducive to the formation of rocky planets. Multiple models have also shown that stars with larger disks are more likely to form terrestrial planets.

Since all of these theories of rocky planet formation begin with the assumption that planets are born from micron-sized dust grains around young stars, finding a persistent disk around a young star can mean finding a habitable planet.

The stars whose NICMOS images were re-examined by Soummer were originally pegged as being good candidates for possessing disks, because infrared telescopes such as IRAS and the Spitzer Space Telescope had found tell-tale signs of warm dusty material associated with each star.  Dusty clouds and disks can be heated by and also scatter light from their stars at wavelengths longer than what can be seen with the naked eye. Special infrared detectors on space telescopes let astronomers observe these wavelengths.

Hubble used its Near Infrared Camera and Multi-Object Spectrometer (NICMOS) to probe these 400 stars for signs of planets in the infrared. NICMOS was the Hubble’s “heat sensor” prior to the newer camera WFC3. It consists of three cameras sitting inside a super-chilled thermos bottle on board the satellite. Installed during a 1997 service mission, NICMOS is a second-generation Hubble instrument that can see faraway galaxies, observe the changes in planetary atmospheres over time, and, most importantly for planetary formation, look through dust clouds.

“All of the images were obtained using coronagraphy,” said Hines, “There is literally a very small pinhole in one of the mirrors inside NICMOS. We point HST such that the light from the star falls within this hole, but all the light surrounding the star doesn’t. This is like putting your thumb over a bright headlight on a car approaching you; your thumb blocks this light allowing you to see the car.”

From 1999-2006 NICMOS scrutinized hundreds of stars selected to be good candidates to host disks or planets. In the end, many of the disks identified by their infrared signals with Spitzer were not able to be seen in scattered light even using Hubble’s NICMOS instrument.

Planets are believed to form and sweep aside the dust in their embryonic disks. Image Credit: The Hubble Space Institute, NASA and STScI

Planets are believed to form and sweep aside the dust in their embryonic disks. Image Credit: The Hubble Space Institute, NASA and STScI

“Even with a coronagraph that blocks most of the direct starlight [sic] there remains a lot of spurious starlight,” said Soummer, “The classical technique was to calibrate out this residual starlight by subtracting a single image of a reference star.” That technique made it impossible to see disks around stars where the disk was too faint compared to the starlight in the image.

Enter Team Soummer. Using software algorithms commonly employed for facial-recognition, noise reduction, and a huge library of reference stars, the archived Hubble images were re-analyzed and re-constructed. It is now possible to see a disk around a star the same way that it’s possible to see a pine tree and a maple tree standing next to each other in a forest. After one expands the computer’s definition of “tree”, the pine and the maple can be distinguished. Likewise, algorithms can help separate the star and the disk in the NICMOS images.

“Previous methods used to analyze these images when they were obtained originally, were fairly rudimentary,” said Hines, “But over the last decade, more advanced techniques have been developed. We were convinced that employing these new techniques, we could reduce the instrumental artifacts that had hidden the disks from us using the older, less robust techniques.”

The re-viewed images revealed not only the debris disks themselves but also their shapes. Of the five disks found, two are laying at angles, so that we can look at them as if we were looking down at plates on a slightly tilted table. Both of these tilted disks have rings, indicating planetary formation.

The other three disks appear edge-on to us. That means that we can’t appreciate any rings that might be present, but we can tell if the edge is wobbly, or asymmetrical, which would indicate planetary formation.

“Now, with such new technologies in image processing, we can go back to the archive and conduct research more precisely than previously possible with NICMOS data,” said Hines.

One of the stars, HD 141943, is almost a carbon-copy of our own Sun during its young, turbulent, planet-forming days. By studying this star, we can understand how our own Solar System swept into existence.

The Subaru Telescope captured this near infrared image of the protoplanetary disk around PDS 70. Credit: NAOJ

The Subaru Telescope captured this near infrared image of the protoplanetary disk around PDS 70. Credit: NAOJ

This discovery, published in Astrophysical Journal Letters, demonstrates that a second-look can be eye-opening. A second-look at these stars increased the number of debris disks seen in scattered light by 21%. Future second-looks may increase the odds of finding planets, formed and unformed, many-fold.

“We have about 400 stars to reanalyze that have been observed through the HST/NICMOS coronagraphic lifetime. So far we have processed about 2/3 of them,” Soummer “In addition to looking for disks, we are also looking for planets and sub-stellar companions.”

Soummer and his team are planning a third look at these same stars, with the Hubble Space Telescope and the Gemini Planet Imager. Several famous exoplanet systems have been inferred to exist by the presence of dust disks. With this new technique in hand, who knows how many more planets may be uncovered, hiding in plain site, in infrared light.

“Just finding more planets around other stars increases the likelihood that there are habitable planets that may contain life, finding more disks, especially those made or rocky stuff, increases this likelihood,” said Hines, “These disk also give us insight into the probable composition of any planets in these systems. Because the dust is easier to “see” that any planets themselves, we can better analyze the material that is being incorporated into planets. [sic] Life as we know it needs the elements SPONCH plus CaFe (sulfur, phosphorous, oxygen, nitrogen, carbon, hydrogen, calcium and iron. By examining disks, we can inventory such elements and better understand the chances that these elements will end up in planets.”


Funding provided by NASA through the HST archival research program and the Sagan Fellowship Program executed by the NASA Exoplanet Science Institute.