Looking for planets is a difficult task. Planets are physically small (compared to stars), physically faint (compared to stars), and are consistently located next really bright objects (those would be the stars). Looking at planets isn’t much different from looking at bats eating bugs in front of the largest spotlight you’ve ever seen. As the bats swarm in and out of the light, they eclipse a small amount of the spot light, and the fluctuations they create can (with a precise enough detector) be measured (this is an indirect bat detection). The bats can also be detected from the light their little black bodies reflect from the spot light out to an observer. Finally, with night vision goggles (and a lot of care not to actually look at the spot light), their little mammal body warmth can also be detected as infrared light.
So, when we look for planets directly, we need instruments that can look for the light the planets reflect from the star they orbit, instruments for detecting the infrared light they give off because they are warm, and design instruments that lest us do all of this while not getting blinded by the pesky central star. In his October 7 paper on arXiv, Steven Beckwith calculates that a space telescope of 10 meters or more (and preferably more like 16m) is required to detect terrestrial planets within the habitable zone of stars. (Fine point here, while the habitable zone moves as stars change in size, this doesn’t help – Big Stars are really bright and can can interfere with things at the larger radii where habitable planets must be located, and little stars have to keep their habitable planets really close in, which makes them hard to see even in the little star’s faint light).
To come up with his 10m and larger number, Beckwith asked, what diameter is needed to resolve the planet from the star when the star is farthest from its star. This number changes with wavelength, and larger telescopes are needed to resolve small features at longer wavelengths (in English, if I want to see details on Jupiter in an optical telescope, I only need something a few inches across, but if I want to resolve those same features with a radio telescope, I need a telescope a billion times bigger. He states, “A spectrum should extend to at least 1 micron to be analyzed for evidence of disequilibrium chemistry indicative of life, and longer wavelengths require lengths bring in even more interesting features.” In other words, to see the neat molecules that result from life (more below) we need to look at infrared light. To resolve 1 micron light from a planet 27 pc away, he estimates an 8-m space telescope is required. To see planets out to 54 pc, a 16-m space telescope is required. For reference, James Webb Space Telescope will be 6.5-m. The larger the telescope, the farther away we can resolve planets, and the farther away we are able to look, the more of space we can sample. Mathematically, this has the neat effect that for every doubling of the telescope’s size – that increase from 8-m to 16-m diameter – the number potential planets we can look for goes up by a factor of 10.
What makes Beckwith’s paper strong is how he details what happens as real world issues are taken into account – things like planets not generally being ideally located at their most distant point on the sky from their star, the effects of background stars, interference from dust and gas in the distant solar system and other annoying side effects.
He also discusses ways to detect planets by blocking out the light of the star instead of trying to resolve the planet in the glare of the star. This sounds like it should be easy – just throw a chronograph (a disc that blocks the star’s light) on the scope and go. The problem is, light waves coming around the disc can interfere with one another and cause diffraction. One possible solution, put forward by Webster Cash and a team at Colorado University, suggests using a flower shaped screen to block the star’s light from a distance. With this technique, a 60m screen 400,000 km away could allow a 4m telescope to efficiently find planets like the 8m mentioned above. This raises the awkward question, is it easier / more cost effective to build 8m telescopes with very precisely built chronographs or 4 m telescopes and 60 m screens that have to be separately controlled?
He also gives a detailed discussion of how planetary atmospheres can be observed in the spectra of their parent stars, making the key point “somewhat non-intuitively … the atmospheric signature for a planet in the habitable zone is independent of the planetâ€šÃ„Ã´s size; … it depends only on the density and temperature” of the atmosphere. In the perfect storm of planets consistently having just the right atmosphere and crossing stars across the middle of their disk, an 8-m telescope will only be able to resolve organically interesting feature in a few (like 3) solar systems in the solar neighborhood. This isn’t good. Beckwith estimates that only 1 in 100 stars searched will have a planet that may have the right characteristics to look for life, and our imaginary 8-m telescope just isn’t likely to let us find it. He estimates that a telescope 16-m in diameter would have a chance to find 1000s of candidate stars and 100s of possible planets to remotely explore for life.
It’s in the second paper, by L. Kaltenegger of Harvard CfA dna F. Selsis of CRAL-ENS (France) that we learn what to look for when searching for life. This team follows up on the 1993 work of Sagan (yes, the Carl Sagan) and collaborators who directed the Galileo Space Probe to turn its detectors back on Earth to see if they could detect life on Earth as the probe flew toward Jupiter. Sagan and company detected molecular oxygen and methane gas in Earth’s atmosphere, as well as a distinctive absorption of red light that indictes biophotosynthesis. In this new work, they point out that in addition to these biomarkers, other gases including water vapor Nitrous Oxide, Hydrogen Sulfide, Carbon Dioxide and can also be used to find life. Methane and N20 are both produced by life on Earth, and N2O in particular isn’t produced by many natural processes. Seeing H20 and C02 can indicate a planet that could have life, and finding N20 probably means life is there.
This raises the question, what is required to find the molecular lines of these molecules? Well, as stated above, a big telescope. The US has proposed the construction of a satellite called Terrestrial Planet Finder (TPF) and Europe is considering a mission named Darwin. TPF will be able to see molecular oxygen and water vapor and Darwin will be able to see molecular oxygen, carbon dioxide and some water lines. It is a start. Unfortunately, many of the biomarkers can be easily lost or confused in the rich forest of molecular lines that appear in gases seeded with Carbon. But, its a start.
Ten years ago we were just starting to find exoplanets. Today we can can count over 200 of them out among the nearby stars. 5 years ago we were just starting to detect them via transits, and today amateur astronomers are in on finding the faint fluctuations that indicate planets. 3 years ago we just started to detect the light of planets separate from the light of stars. Each time, the first discovery has been a delightful surprise. While this is really hard work, technological breakthroughs keep surprising us and making it easier to find new worlds. We know what we need to look, in terms of technology, and we know what to look for in terms of molecular lines in spectra. Now, we just need the money to build the instruments that will allow us to go forward and discover.
Hat tip to Fraser Cain for pointing out these two great papers.