Detecting extra-solar earths: NASA's Terrestrial Planet Finder mission
"I should disclose and publish to the world the occasion of discovering and observing four planets, never seen from the beginning of the world up to our own times, their positions, and the observations...about their movements and their changes of magnitude; and I summon all astronomers to apply themselves to examine and determine their periodic times..."
- Galileo Galilei, March 1610
by Ruslan Belikov
Thus spoke Galileo Galilei, the first person in history to point a telescope up at the heavens and record his observations. The four planets he spoke about are the four largest moons of Jupiter, subsequently named Galilean moons. The impact of seeing these objects was greater than this quote alone would suggest. Galileo's observation has dethroned the Earth from being unique: here are, for the first time in history, heavenly bodies that are manifestly orbiting something other than the Earth, which was held to be the center of the universe in those times.
Fast-forward four centuries.
Since the humble beginnings of telescopic astronomy in Galileo's hands, humanity has built telescopes 10 meters in diameter on the ground and put several smaller ones in space. Not only has the Earth been dethroned as being the center of the universe, but so has the sun, and the Milky Way galaxy. In fact, the very matter that comprises us and everything we observe is not even the main "stuff" of the universe - the majority is the puzzling "dark matter" and "dark energy."
Surprisingly, one of the last things that was dethroned is the uniqueness of our sun's planetary system. To be sure, there was little doubt other stars have planets, but there was no factual proof of this until quite recently. The first definitive extra-solar planet (around a main-sequence star) was announced only in 1995 by Michel Mayor and Didier Queloz, University of Geneva. Since then, the last decade has seen an explosion of new extra-solar planet discoveries: over 150 have been found to date.
However, all these planets are big, Jupiter-like gas giants (excepting perhaps some pulsar planets and one very recent find), necessarily so because humanity's feeble instruments can only detect the largest of worlds. These behemoths are incapable of sustaining life as we know it and seem just as alien and barren to us as our own Jupiter. The Earth is still the center of the universe in the sense that, as far as we know, it is the center of life. At present, any analogue to Earth can be found only in the pages of science fiction. The ultimate discovery of another Earth-like world would herald a new era, an era in which Earth-like planets, be they barren or teeming with life, are known to exist elsewhere in the universe.
Such a discovery may very well take place as early as 2015. If funding persists and everything goes on schedule, that year will see the launch of a NASA space telescope called the Terrestrial Planet Finder Coronagraph (TPF-C). This telescope will survey the nearest few hundred stars that are most likely to harbor an Earth-like planet. It is unlikely that we will actually find an advanced civilization on those neighboring worlds (if one existed, we would surely have picked up their radio broadcasts by now), but that is not the goal of the mission. The primary goal is to find planets, life or no life, so that we can learn more about our own Earth, just as a psychologist needs to study many people, big and small, young and old, man and woman, before he can truly understand one. How common are Earth-like planets? How do they form and evolve? How diverse are they? Do they harbor the conditions for life? Is life unique? If not, how common is it, how diverse, and how does it form? Ultimately, where did we come from? These are the questions we hope to shed light on with TPF-C, appropriately a part of NASA's Origins program. The price of getting those answers? An estimated $2 billion. An eyebrow-raising number, but yet it is less than a dollar per year per American for the next 10 years. Think about that the next time you buy lunch!
How can we answer these questions simply by looking at an image of a planet? To make matters worse, we will not even get a resolved image of a planet, just a single blurry speck, the best our feeble instruments could do. Nonetheless, there is a wealth of information that can be teased out of that single speck. One obvious parameter we will be able to measure is its brightness, from which we can infer an estimate of the planet's size. We can also measure the orbit and distance from the star, which would let us estimate the average temperature of the planet. If different sides of the planet reflect different amounts of light (as they do on Earth due to the varying distribution of landmass), we will be able to measure periodic variations in brightness as the planet rotates and thus measure the length of the day.
A wealth of further information can be revealed from the spectrum of that single speck, such as the presence of the atmosphere and its pressure, as well as the abundance of various compounds such as oxygen and water. If there is plant life and if it is anything like that on Earth, it will manifest itself via a characteristic increase in reflectivity at infrared wavelengths called the "red edge." In short, we will be far from bored with that one speck.
However, the actual detection of an Earth-like world is very difficult. There are two reasons for this. The main reason is that the star is many times brighter than the planet. As viewed from far away, our sun would be almost 1010, or 10 billion times brighter than Earth. This is about as bright as a powerful searchlight would appear next to a firefly. The second reason is that the stars are so far away, so that one needs very fine resolving power in a telescope. From one of our neighboring stars (say, 10 parsecs away), our Earth would appear only 0.1 arc seconds, or 0.0015 degrees away from the sun. This would be equivalent to trying to see the firefly buzzing within 6 feet of our searchlight while staring directly at it from a distance of 2,600 miles, or roughly New York to Los Angeles. As hard as it is to imagine, there are already instruments that meet each of the two requirements separately. Meeting these two requirements simultaneously is the main technological challenge of the mission.
The terrestrial planet finder mission TPF-C is being pursued by NASA's Jet Propulsion Laboratory (JPL) in Pasadena, Calif., along with a few subcontracted teams in academia and industry, our team at Princeton University being one of them. The TPF-C telescope will have an elliptical primary mirror that is roughly 8 by 3 meters, which provides the necessary resolving power, and will image in visible and near-infrared wavelengths. It will be placed in space to eliminate the effects of atmospheric turbulence. (The reason the mirror is elliptical is so it would fit into the launch rocket!)
However, even though the large mirror size lends the required resolution, a conventional telescope design will not provide the required 1010 contrast. The reason for this is that in conventional telescopes, the star image is not a tightly confined dot or a circle, but, well, a "star" shape: it is an extended pattern with glare and perhaps four or more streaks. An object not much dimmer than the star could be seen through this glare, but a planet that is 1010 times dimmer will be completely obscured. In order to reduce the glare, the back of the telescope will contain a special high-contrast system called the coronagraph (so called because these were initially conceived for observations of our sun's corona).
However, conventional coronagraphs are not powerful enough to achieve 1010 contrast, and new designs are being developed. One promising design is the Shaped-Pupil Coronagraph being pursued by our group at Princeton University. The basic principle is this: the image of a star in a telescope is the so-called Point Spread Function (PSF), which is the Fourier Transform (FT) of the telescope opening, the pupil. Typically, the PSF is not tightly localized, causing glare. The idea behind our Shaped-Pupil Coronagraph is to shape the telescope opening, or pupil, so that its PSF is tightly localized, providing 1010 contrast in the desired regions around the star where a planet may reside.
It turns out that in order to maintain the 1010 contrast provided by the coronagraph, all the telescope mirror surfaces need to be precise to at least 1 angstrom. That is less than the size of an atom! Furthermore, the reflectivity uniformity of the mirrors has to be better than 1 part in 1,000. The state-of-the-art mirrors today can only achieve surface variations of about 100 angstroms and reflectivity uniformity of only about 1 part in 100.
In order to bring these figures down to the required levels, corrective so-called Extreme Adaptive Optics (EAO) systems are being developed. These rely on so-called deformable mirrors (DMs), or mirrors whose surface can be actively controlled, to precisely cancel out the aberrations of all the optics in the telescope. The JPL team headed by John Trauger has demonstrated a 109 contrast after EAO corrections (albeit for just one wavelength) for its type of coronagraph. As of this writing, our Princeton group has not yet tested our Shaped-Pupil Coronagraph with an EAO system, but we are getting 105 to 108 contrast (depending on the distance from the star) before any corrections. Many challenges remain, such as how to control for wavelength-dependent aberrations, but we feel confident they can be resolved.
The Terrestrial Planet Finder Coronagraph is surely to be but a first of many future missions to detect and study Earth-like planets. A follow-up mission, TPF-I (I standing for Interferometer) is being planned for a launch in 2020 and will conduct further science on Earth-like planets in the infrared. Eventually, technology will advance enough for us to be able to resolve features on planets, and several (still prohibitively expensive) missions have been conceived to do just that.
Alas, vast interstellar distances will prevent anyone from making the journey to any extra-solar planets in the foreseeable future, at least not within a single lifetime. However, there is one thing we can foresee with almost certainty. In 10 or so years, Earth will have gained a sibling, "never seen from the beginning of the world up to our own times."
Dr. Ruslan Belikov was born in Kyiv and came to the United States in 1991. He received his Ph.D. from Stanford Univeristy in 2004, and was awarded the Michelson Fellowship in 2005 from NASA's Jet Propulsion Laboratory to conduct research at Princeton University's Terrestrial Planet Finder Laboratory. This article is based on a lecture he delivered recently at a meeting of the Mathematics-Physics-Technical Section of the Shevchenko Scientific Society, of which he is a member. He may be reached at email@example.com. For more information on TPF, please go to http://www.princeton.edu/~tpf/ or http://planetquest.jpl.nasa.gov/TPF/tpf_index.html.
Copyright © The Ukrainian Weekly, July 10, 2005, No. 28, Vol. LXXIII
| Home Page |