TESS, NASA’s new space telescope for detecting planetary transit events was successfully launched by a SpaceX Falcon9 rocket today, with the first stage successfully landing on its recovery barge for re-use.
Everyone familiar with FtB should also be familiar with Kepler, both the rabble-rousing white dude and the chrome-hot, super-cool, USD $700 million, solar powered camera. Kepler was initially expected to generate data for no more than 3.5 years, though it has exceeded that by a wide margin despite failures to its gyroscopic stability controls that were originally thought mandatory to complete the mission. It is true, however, that the mission’s data collection is much slower now.
Enter TESS: The Transiting Exoplanet Survey Satellite. TESS is a next-generation solar powered black tube with a mirror and a camera, and dang is it fine. Estimates of the number of exoplanets Kepler might find were rarely very precise, but its 2652 confirmed planets*1 and 2724 yet-to-be-confirmed planet candidates*1 clearly more than satisfied expectations for the telescope. But even cooler than that, the planets found compared with the known number of stars in the Kepler camera’s field of view allows much better and more precise estimates of TESS’s expected future discoveries. While depending on confirmation rate Kepler will ultimately be credited with 4000-5000 confirmed planetary discoveries, TESS should rack up many, many more if its components last for their expected lifetimes:
“The number of Earth-sized and super-earth planets that TESS should find over the course of its two-year primary mission will be in the range of 500 to 1,000 new planets, and overall the number of planets that will be established is likely to be in excess of 20,000 all together,” enthuses MIT’sGeorge Ricker, who is TESS’ Principal Investigator.
Note that “earth sized and super-earth planets” is probably an informal reference to planets of approximately earth-mass (at least 1/3rd of earth mass, with an upper bound varying depending on the source, but frequently approximately 2-3 earth masses) plus planets categorized as “super-earths” (having a mass exceeding that of an earth-mass planet, whose upper bound can vary as noted, but not exceeding 10 earth masses, which limit has more general agreement than the earth/super-earth mass dividing line).
Though Ricker obviously knows a heck of a lot more than I do, that number seems conservative to me. TESS has more sophisticated systems to measure brightness changes, and this Caltech archive notes that of the 734 confirmed planets discovered by Kepler and whose mass was later successfully measured, only 121 were determined to be of a mass placing them in the earth-mass or super-earth categories (caltech used 3x earth mass as the dividing line between earth and super-earth). Since only about 1 in 3.5 planets were successfully weighed, we should expect 20k/3.5 mass-evaluated planets from TESS, with a similar percentage of planetary masses in the earth/super-earth range giving us an estimate of somewhere between 900 and 950 planets. With TESS’s increased ability to detect smaller planets, it’s possible we could see many more than 1000 earth/super-earth planets being discovered on this new mission.
If I had to guess, not being a planetary scientist or any kind of astronomer, I’d say uncertainty over length of mission is responsible for the conservatism. As much as sensitivity, length of mission has the capacity to affect the number of discoveries. Remember that to discover a planet candidate requires 3 observed transits with two (nearly) identical periods between. To discover earth via a TESS- or Kepler-like mission orbiting a star in our neighborhood, even with more than enough sensitivity it would require a full 2 years of observations for earth to register as a candidate planet even if it happened to transit the first day of that telescope’s operation. If earth had just transited the day before the telescope began operating, it would require 3 years just for earth to get any initial notice.
Because of the unpredictability of space operations, TESS’s primary mission is scheduled to run for a mere 2 years. For solar systems like ours, that’s only enough to note the existence of Mercury and Venus, and the sensitivity may not even be enough to notice Mercury. Given that, our solar system to a TESS-like mission would appear to have only a single planet after two years. But if there are no mishaps, and if TESS holds up at least as well as Kepler, we could see a mission length long enough for our imaginary extra-solar system TESS-analogue to easily discover earth and even possibly to discover Mars. This doubles or triples the number of discoveries at the same sensitivity. While it’s still not possible to know how many stars have solar systems similar to ours (sensitivity hasn’t been good enough to detect a Mercury or even a Mars before TESS, and we haven’t been observing long enough to detect any Jupiters or longer-period planets), there are a huge number of planetary systems with a star’s satellites concentrated in close-in orbits, and so doubling or tripling mission length might generate less than double or triple the number of planetary candidates. On the other hand, if it turns out longer-period planets are ultimately more common, if mean-period is optimal a doubling or tripling of observation time could yield more than a doubling or tripling of candidates. We’ll have to wait and see.
In the meantime, let’s just be excited that this amazingly cool tube with a black interior, a mirror, a camera, and a few solar panels is now outside earth’s atmosphere and ready to push itself into its observational orbit.
*1: These numbers include all confirmed and candidate planets found using the Kepler instrument, though for reasons having to do with budget, the Kepler mission was shut down after the instrument was reduced to 2 gyroscopes. When a proposal proved viable to use innovative stabilization strategies to enable renewed data collection, congress granted money for a new mission using the same Kepler telescope. The new mission is named K2. In most venues, confirmed planets found during the Kepler instrument’s main mission – a mission also called “Kepler” – are tracked separately from those found during the K2 mission, despite the use of the same space hardware. I added available numbers together as I thought for most persons the Kepler/K2 distinction was largely irrelevant.
Wonderful to hear about TESS! Note that TESS’ observing strategy is fairly different from Kepler’s. TESS’ camera is less precise but it will search for planets around many more stars. While Kepler stared at the same stars for years, TESS will search a new group of stars every month (though some parts of the sky will be monitored for longer). So, TESS will search many more stars, but for less time and not as precisely. That means it will find lots and lots of planets very close to their stars (so the orbital periods are short, generally shorter than about 10 days) and that its discoveries will be mostly around bright stars (which are easier to measure). Those two factors, short orbital periods and bright targets, will make these planets much easier to observe from the ground. TESS should keep the Earth-bound telescopes very busy trying to measure masses for these planets and maybe even trying to detect their atmospheres.
Oh, hey, I didn’t realize that it was going to be switching sky patches quite that often. Thanks for the info!
Too bad JWST is going to have too many missions to really do much planet hunting.
Yes, JWST won’t be doing much planet hunting (it wasn’t designed for that). But it will spend time looking for atmospheres on planets discovered by other telescopes including TESS. That’s a big part of the idea behind TESS’ strategy: to tell the other telescopes where to look.
Marvellous news and blog post here thanks. Great to see TESS fly and can’t wait til we get to know more new worlds round other stars from this.
Not sure of the target stars and it could vary but the majority of stars out there* are actually fainter and smaller and dimmer than the sun – to be astronomically pedantic here classes K & M orange and red dwarfs and dimmer, cooler G type main -sequnce aka dwarf stars. Now this means that planets orbiting closer in to those stars for instance as close as Mercury or closer can be cooler and thus potentially more habitable despite having shorter orbits i.e. years.
There are complicating factors like flares, tidal locking and soon but generally there’s hope that for many systems planets orbiting where, say, Venus** is may actually be habitable and worlds further out too cold to sustain life although as Europa and Enceladus may show perhaps the closeness to suns isn’t the defining factor in whether life can exist anyhow.
* Spectral classes by percentage of numbers present apparently breaks down as :
70% red dwarf stars. (Type M,cool, dim, not one visible to the unaided human eye, often flare dramatically and likely to have planets inhabitable zone tidally locked so problematic for habitable worlds but very long lived).
15% orange dwarf stars (type K, cooler and fainter than our Sun, a few visible to the unaided eye e.g. Epsilon Eridani, Epsilon Indi, 61 Cygni,)
10% white dwarf stars – or really ex-stars these are the cooling core remains of stars that that have used up all their fuel and aren’t actually generating energy. The size of Earth and initially blue hot they take aeons to cool off but likely don’t have any planets orbiting them closely enough to be habitable since they would have destroyed close in planets during the red giant stage of their lives.
4% G type stars – like our Sun which is actually at the bright end of the G class although Alpha Centauri A is brighter than it, Tau Ceti and 55 Cancris are other examples that have planetary systems but are dimmer than our sun.
1% A & F stars ranging from a bit hotter and brighter and more massive than our sun to double our suns mass and twenty plus times for the massive end of this category -stars like Sirius the Dogstar, Procyon the Little Dogstar, Vega, Altair and Upsilon Andromedae. Despite being main-sequence stars they’re not usually called “(yellow-)white dwarf” stars because of possible confusion with white dwarfs which are ex-stars -see above! These stars are shorter lived than the truly sun-like stars meaning at the more massive end they probably don’t live long enough to evolve advanced multicelluar life.
All other types of stars make up less than 1% including alltheginats , supergiants and O-& B type blue “dwarfs”which are all much brighter than our Sun and make up most of the visible stars in our sky despite being very rare stars because they shine so brightly they can be seen for hundreds sometimes thousands of light-years away.
Source for stellar demographics here is Ken Croswell’s P. 31 diagram Is there life around Alpha Centauri” in article in Astronomy </imagazine April 1991.
** Bit of a tangent here but have you read Phil Plait’s article here suggesting there may well be life in the clouds of Venus? :
http://www.syfy.com/syfywire/life-in-hell-could-venus-have-a-bacterial-infection
PS. Sorry about my italics fail there, mea culpa. Oh & all of my typos too, apologies and hope y’all follow my gist there anyway.
NB. that 1991 Ken Croswell article was written before brown dwarf stars classes L & T & Y were added or discovered so ignores them. These brown dwarfs are likely fairly abundant and numerous but probably not the best hosts of planets for life – extremely cool, some notably active flares~wise and even more likely tidal locking so that one side of the world always faces its host and is permanently baked and the other never does and is consequently frozen – although, again, there are a number of possible complicating factors and so much we don’t know especially about life and its varieties and possibilities so.. Anyhow, again, great news and blog post here.