Help - Search - Members - Calendar
Full Version: Kepler Mission
Unmanned Spaceflight.com > Beyond.... > Telescopic Observations
Pages: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25
brellis
Thing that gets me is, the nominal mission duration is 3.5 years. Can we puhleeeeze extend it so we can nail down some more earth-type planets?
ZLD
Didn't notice this til far too late in the day but for anyone interested, there was an impromptu, 'Ask Me Anything' (AMA) interview on Reddit today from a Kepler mission operator (unsure of name) for anyone interested. Link
djellison
QUOTE (brellis @ Feb 2 2011, 10:25 PM) *
Can we puhleeeeze extend it so we can nail down some more earth-type planets?


Let's see if they apply for one, and if their plan is fiscally responsible, and then they might get one.

It's too premature to be talking about it now.
cassioli
QUOTE (Hungry4info @ Feb 3 2011, 12:36 AM) *
So yes, data is still being analysed. As awesome as this announcement is, it is still the beginning.

Have you got any source link I can put on my blog?
I'dl like Kepler team was more clear about which data it's releasing. This page, for example, is very confusing:
http://archive.stsci.edu/kepler/data_release.html

Does it exist any statistic about percentage of data analyzed, and "level of confidence" that results on a candidate planet (being them positive or not) ar final?
I mean, they talk about 15 total confirmed planets: how many candidates have been deeply analyzed, and "how much" deeply, to get this result? 20 candidates? 200 candidates? All 700?

About patience: I know it's needed evry every time you talk about space-related matters, but... simply I don't have it! rolleyes.gif[
djellison
QUOTE (cassioli @ Feb 2 2011, 11:30 PM) *
I don't have it! rolleyes.gif


Well, tough. You have to be patient. Fact.
ZLD
The mission has already lasted almost 2 years, and with an expected 3.5 years, that really isn't a lot of time left before the minimum primary mission conditions are met. That is still long before we see much from Rosetta, Juno, New Horizons, Dawn at Ceres, and Akatsuki's second attempt at Venus. There is a lot to look forward to in space science in the next couple of years. Before long, you'll be overwhelmed by data, so sit tight.
peter59
I just read an article about the Kepler-11.
Kepler finds a mini solar system!
Makes me laugh so accurate publication of the size and mass, for planets in case when is not certain whether they really exists. It is rather a quackery, not real science. Published sporadically over the past thirty years, "discovery" of extrasolar planets do not excite me completely, always in the end it turned out that their findings are questionable. I'm very skeptical, we need a much better tool than Kepler.

ZLD
People have been observing transits on Earth for quite some time so the science is already decently established and checked against what we already know. It's actually reasonably easy to extrapolate information such as mass, by knowing just the duration of the orbit and how much the star dims. The numbers are rough estimates for sure, but certainly not quakery.
ugordan
QUOTE (peter59 @ Feb 3 2011, 03:06 PM) *
I just read an article about the Kepler-11.
Kepler finds a mini solar system!
Makes me laugh so accurate publication of the size and mass, for planets in case when is not certain whether they really exists. It is rather a quackery, not real science.

I disagree. The graph shown obviously illustrates the uncertainties in planet radii and masses, the reported sizes are probably just the most likely numbers. They have large error bars, but it's still most-probable figures. I don't really expect a popular science article to state error bars with 3 sigma certainty levels and such.

As to whether the 6 planets really exist, I haven't read the paper, but from what I've seen in the press conference it's basically a done deal. The mass information in particular came from transit timing variations on the order of about 10 minutes caused by gravitational tugging of the inner planets. This effect is only observed in an actual multi-body system.
FatSplenda
QUOTE (peter59 @ Feb 3 2011, 06:06 AM) *
I just read an article about the Kepler-11.
Kepler finds a mini solar system!
Makes me laugh so accurate publication of the size and mass, for planets in case when is not certain whether they really exists. It is rather a quackery, not real science. Published sporadically over the past thirty years, "discovery" of extrasolar planets do not excite me completely, always in the end it turned out that their findings are questionable. I'm very skeptical, we need a much better tool than Kepler.

Clearly you misunderstand... everything. I hope you're a troll.

First, there weren't any publications of discovered ("discovered," for you) exoplanets until about 15 years ago.

Second, did you ignore their error bars? Their statements about statistical uncertainty? Their admitted lack of knowledge in some places? It's all over their papers. Kepler science is more about statistical variation and analysis than just about anything else. They place extremely tight statistical constraints on top of an already reluctant attitude. The biggest skeptics of Kepler are those who are doing its science.

I'm not sure why you're calling it "not real science." What in particular? What are your improvements? Where do you derive your skepticism from? What makes exoplanet discoveries "questionable"? Why isn't Kepler suitable? What do you know about transit timing variation methods? What do you know about the spacecraft attitude constraints and abilities? What do you know about CCD signal-to-noise ratio? I'm ready and willing to hear.

It seems you just don't like when they quote a number to two decimal places (nevermind the uncertainty on it!) on something that hasn't yet been confirmed. Would you rather have it the other way? "We have 1235 signatures of planets, but because they aren't confirmed planets yet, we won't tell you the numbers."

Don't not believe in something because you yourself couldn't do it, or because you don't have insight into the extremely complex system that you're criticizing.

EDIT:

QUOTE (brellis @ Feb 2 2011, 10:25 PM) *
Thing that gets me is, the nominal mission duration is 3.5 years. Can we puhleeeeze extend it so we can nail down some more earth-type planets?

Oh, most definitely! The extension will come easily. If she keeps up this kind of science return for the rest of those 3.5 years, you can bet they'll be pumping money into extending it and its follow-on missions.

It's still amazing to me that when I say "this kind of science return," I'm talking about only 3 months of data! We're coming up on 2 years of mission time here, soon!

(Ultimately, it's about resources. When I worked on the mission, there was enough fuel on board (for momentum desaturation maneuvers) for about 5 years. They adjusted their algorithm for computing thruster-on times to improve this number to beyond 6 years.)

QUOTE (ZLD @ Feb 2 2011, 10:36 PM) *
Didn't notice this til far too late in the day but for anyone interested, there was an impromptu, 'Ask Me Anything' (AMA) interview on Reddit today from a Kepler mission operator (unsure of name) for anyone interested. Link

Hey, yeah! I worked with him when I was at LASP! Chatted with him on Gchat last night. His name's Jake.

Keep in mind that he's a student, too; so was I, when I worked on Kepler. The real-time ops and first-order engineering health and safety analysis is done by people with no degrees!
Drkskywxlt
As far as confirming Kepler planet candidates, look at the post I made a few pages ago summarizing the presentation by the Kepler RV follow-up team at the last meeting of the ExoPAG. They will "only" be able to confirm ~25 potential Earth-sized planets and those will likely be in short orbits around small stars. This of course assumes nominal mission funding with no extended mission. Kepler and the potential gravitational microlensing missions (WFIRST) will mostly tell us information about statistics...how many stars have planets, of which sizes, at which orbits, etc. Most of the Kepler candidates and all the microlensing candidates will never be confirmed with RV, astrometry or direct imaging. They're just too far away around stars that are too dim.
Hungry4info
QUOTE ("cassioli")
QUOTE ("Hungry4info")
So yes, data is still being analysed. As awesome as this announcement is, it is still the beginning.
Have you got any source link I can put on my blog?


15 < 1,235

These two papers make it clear that we're not done yet with the follow-up confirmation process. Note the presence of more than a thousand candidates awaiting confirmation, and many more multi-planet systems, as Mongo pointed out.

http://arxiv.org/abs/1102.0543
http://arxiv.org/abs/1102.0541
Explorer1
I'm a regular reader, and the Bad Astronomer's blog is the last place to be complaining about 'quackery'. He has an account on this forum as a matter of fact.

Like ugordan said, the six planet system is confirmed, the other 1350 or so are the ones not yet verified. The chance that they're all false positives is vanishingly small.
Marz
This is hugely exciting! I realize more measurements are required to confirm many of the candidates, but the results look very promising.

I wonder if these results will rekindle interest in the Terrestrial Planet Finder and/or Darwin? As far as I know, those projects are mothballed, so is there anything else in the works?

Would follow-up Keplerish missions be worthwhile, or is the objective of Kepler mostly to characterize solar systems and not find new planets? Is it possible to design a Kepler II scope that can survey several rich field-of-views before mission end?

I recall reading a paper about the concept that stars may reveal clues to solar system composition in their spectra - could this be used as a preliminary means to determine the prevalence of elements in the system?

Would any of this data be useful to SETI to focus monitoring on areas of the sky with known potential?

Wahoo!
middleschoolsteve
I'm sorry if I'm missing something obvious; I'm thinking about this while teaching sixth grade science. If the chances of finding an "earth" orbiting a "sun" using the transit method is about 0.5%, because of the alignment problems, don't the 54 candidate planets imply that there is probably a huge total number of habitable zone planets in the field if we include the planets that do not transit?
I realize that many of the candidate habitable zone planets are larger than earth, so the number isn't 11,000, but it is big.

Even with less density of stars over the whole sky, the numbers are breathtaking.

Then when you think about the rest of the galaxy....
hendric
Middleschoolsteve,
Yes, this is exactly what Kepler gives us: meaningful statistics that aren't as skewed towards large planets as an RV search. I'd say your 10,800 (54/.005) is fairly accurate. (Fairly accurate in astronomy is within an order of magnitude, normally. smile.gif ) So taking 10,800 out of 150,000 candidate stars gives us about 7% chance of planets in the habitable zone. I think the statistics for earth-sized planets in the habitable zone (4) is a little small to make predictions like this across the whole field; I'd prefer to have at least 20 before we make predictions.

The future benefit of Kepler data is not necessarily that telescopes will look at these planets, most are too far away to be seen in even the wildest dreams of astronomers. The benefit is that we can start concentrating our searches on nearby stars that match statisticially with Kepler star groups that have a high percentage of habitable zone planets. For example, metallicity or size groupings. Also, the data will help inform our theories on solar system development; concrete information on multiple planet systems and their stars' characteristics will help improve our models for forming solar systems.
cassioli
QUOTE (Hungry4info @ Feb 3 2011, 06:10 PM) *
Have you got any source link I can put on my blog?

15 < 1,235

I think this is a common mistake: being only 15 candidate planets confirmed as planets does not mean at all that only 15 candidates have been examined: indeed, I read that some hundreds of false positives have been confirmed!
The origin of this uncertainity about candidates/confirmed is that each data release overlaps previous ones, i.e. provides confirmed planets, confirmed false positives, and additional candidates, all together.
Mongo
The obvious next step after Kepler is TESS, which is a similar transit-detecting mission to Kepler, but focused on brighter stars located over much of the entire sky, instead of fainter stars in a small patch of sky. The big advantage would be that planet candidates would be far easier to study, since they would be transiting stars with much greater apparent brightnesses. TESS had been deselected, but indications are that it might be on the table again.

For more background information about the remarkable co-orbital planets at KOI 730, see this paper, written by Greg Laughlin and John Chambers in 2002. They identify three distinct configurations of co-orbital systems: the first consist of the two orbiting objects undergoing tadpole-like librations around an equilateral triangle with the primary (like the Jupiter Trojans with Jupiter and the Sun). This appears to be what we see at KOI 730. The second configuration consists of the horseshoe-type orbits, where the two orbiting objects' librating motions are symmetrical around a 180 degree corotating frame. The Janus-Epimetheus pair is an example of this. The third configuration is an exotic situation where the two co-orbital objects exchange angular momentum in a way that prevents close encounters.

Greg's current Systemic blog post talks a bit about KOI 730 in relation to his earlier paper.
ZLD
My hope is set on Giant Magellan or the ELT being fully funded and built in the next decade to provide us with some better visual confirmation than can currently be performed. The TPF would be far better in this area though and would be much preferable if the Fed was serious about the future of observatories (haha).
hendric
Found a nice place to download the Kepler data in a more-friendly format than FITS:

http://coolwiki.ipac.caltech.edu/index.php..._and_CoRoT_data

http://nsted.ipac.caltech.edu/applications...pler_index.html

Sample output

http://nsted.ipac.caltech.edu/cgi-bin/Time...que_id=20117189

The nice thing is that they output ASCII that can be easily imported into Excel!

Raw data from above minus a linear trend
Click to view attachment
cassioli
QUOTE (hendric @ Feb 3 2011, 09:33 PM) *
Found a nice place to download the Kepler data in a more-friendly format than FITS:

cool finding!
Stu
Have to be honest, I'm a bit disappointed and baffled by some of this negativitry and cynicism towards Kepler. I don't know what more people want. rolleyes.gif Thanks to Kepler, and the amazing science team behind it, we can all, after centuries of wondering and speculating and dreaming, go outside on the next clear night, look to the west and see Deneb standing there above the trees, with Vega gleaming to its right and know that in the unremarkable-looking patch of sky between them there are at least, at least, 15 alien planets orbiting some of those spilled salt stars. And although we absolutely don't know it for a fact, we know, thanks to Kepler, that there's a chance that many, many more of those stars might have worlds orbiting them too, including some perhaps the size of Earth, with some of those possibly orbiting within their sun's habitable zone.

I know. Chance... might... possibly... Not exactly conclusive, is it? But it's a start.

If we don't look we are guaranteed to find nothing. Kepler is looking, and seeing incredible things, with many more incredible things to come, I'm sure.

We've pondered this for centuries, as a species, and I think it's a pretty safe bet that most of us here on this forum, at some point in our lives, have looked up at the sky, from a lonely hilltop, or a quiet beach, or just from our own back gardens, and wondered if there were worlds whirling around any of those distant suns too. Now we know - there are. We know that one of those stars, even if it's invisible to our naked eyes, has six worlds spinning around it. Six! Doesn't that make the hairs on the back of your neck stand up?

The "alien solar systems" we've all grown up with, on sci-fi shows like Star Trek, Babylon 5, Lost in Space, Dr Who, or on more factual programmes like Cosmos, Horizon, or whatever, have all been completely made up, invented, the products of great imaginations, the children of people with a passion for the amazing possibilities that exist Out There. Today we can actually study real alien solar systems, take images of planets waltzing around faraway stars. Before Kepler we were hesitantly dipping our toes in the cosmic ocean. Kepler is going to bring the waters of that ocean roaring up to us and over our feet and up to our knees, pour planets on us like a summer storm.

Yes, these observations and results are quite speculative. The Kepler team are very, VERY careful to use the term "candidates". But whether the Kepler FOV turns out to contain the frothy surf of thousands of worlds suggested by the most optimistic Kepler supporters, or 'just' a few hundred worlds, or even a just the fifteen we know about today, scattered across it like beads from a snapped necklace, that's an incredible thing.

Rejoice in it! smile.gif
Littau
Forgive me if this has been mentioned before, but I can't seem to find any detailed reference on the analysis of this released data and what bounds it puts on the frequency of planets and planetary systems. I imagine the Kepler team has performed this, but perhaps they hesitate to release it at this early stage?
ngunn
QUOTE (Stu @ Feb 3 2011, 08:50 PM) *
I'm a bit disappointed and baffled by some of this negativity and cynicism towards Kepler.


Me too. Where's it coming from? What do they want?

[quote] Rejoice in it! [quote]

Indeed.
Superstring
Just for fun, here is some quick and dirty math here based on a very small sample size (but hey, for the first time we actually have a sample size > 1!):

Kepler has found 5 Earth-size candidate planets in the habitable zone. The Kepler team estimates that about 80% of the discovered candidate planets will be confirmed, so I'll estimate that 4 of these are real.

Kepler's methods of detection only allow it to catch planets that are exactly aligned on the same plane as us. So it is estimated that for every planet Kepler detects, another 200 go undetected.

4 * 200 = ~ 800 Earth-size planets in the habitable zone in the small fraction of the Milky Way galaxy that Kepler is observing. Kepler is observing about 156,000 stars. Since the Milky Way has approximately 200-400 billion stars, we can multiply the previous estimate of planets by about 2 million.

800 * 2 million = ~ 1.6 billion Earth-size planets in the habitable zone in the Milky Way!

Now, that doesn't mean that all of these are necessarily Earth-like. For instance, Venus and Mars also reside in the habitable zone. If our solar system is any indication, perhaps we can estimate that 1/3 of these planets *are* Earth-like to a greater degree than Venus and Mars. That still gives us ~ 500 million Earth-like planets in the Milky Way.

And that's not even considering moons of large gas giants in the habitable zone, much less the hundreds of billions of other galaxies in the universe.

Again, these are all very rough estimations based on a still small sample set. But I couldn't help myself.
Drkskywxlt
QUOTE (Littau @ Feb 3 2011, 05:39 PM) *
Forgive me if this has been mentioned before, but I can't seem to find any detailed reference on the analysis of this released data and what bounds it puts on the frequency of planets and planetary systems. I imagine the Kepler team has performed this, but perhaps they hesitate to release it at this early stage?


Check Arxiv today. The Kepler team published a paper that offers some of this statistical work that covers their first 4 months of observations.

As far as follow-on work...I've heard the Kepler team has an idea for a follow on spectrographic mission that would characterize some planets, but I think the design is extremely preliminary. Not sure if it's even Discovery-class or something more. The Astrophysics Decadal Survey released their results recently and that basically sets the road map for the next 10-12 years. Due to JWST's financial black hole syndrome, there's no big exoplanet mission slated for this time. The big mission, WFIRST, would be partially exoplanet focused (mostly dark energy) using gravitational microlensing to do more statistical work. TPF/Darwin is kicked down the road until the next Decadal Survey...but this decadal did recommend a strong program of technology development to get ready for a TPF-like mission. They also recommended continued ground-based RV/transit studies. Best bet for a big exoplanet mission would be from ESA...Plato. Plato's in the M-class competition with Euclid and Solar Orbiter and would launch in 2017-18 if selected.
Drkskywxlt
QUOTE (Superstring @ Feb 3 2011, 08:07 PM) *
Venus and Mars also reside in the habitable zone.


Venus is not in the habitable zone. Mars is. Venus might have been very early in the Solar System's history.
Superstring
QUOTE (Drkskywxlt @ Feb 4 2011, 01:25 AM) *
Venus is not in the habitable zone. Mars is. Venus might have been very early in the Solar System's history.


Thanks for the correction. Venus indeed resides just a hair outside the habitable zone if we accept 0.725 AU as the inner boundary. Venus's semi-major axis is 0.723 AU.

For simplicity and in an attempt to be conservative, I'll still stand by my estimate of 1/3 of Earth-size planets in the habitable zone actually having liquid surface water. That's obviously speculative though.

Also, I'll amend my previous post to account for the "galactic habitable zone" which contains about 10% of the stars in the Milky Way. This covers the stars that are not too close to the center (where any planets would be more at risk of dangerous radiation) or too far away from the center (where there is a depletion of heavy elements needed for life and rocky planet formation). There very well may be abundant life outside the galactic habitable zone, but just to be safe, it certainly doesn't hurt to factor it into the math.

Taking 10% of the previous estimate = ~ 160 million Earth-size planets in the habitable zone of stars that are in the Milky Way's own habitable zone. If 1/3 of these have liquid water on their surface, that's still over 50 million.
djellison
I would ask all contributors to this thread to re-read the forum guidelines : http://www.unmannedspaceflight.com/index.php?act=boardrules : in particular regarding Exobiology and SETI.

A few private messages have had to be sent as several posts or bumping up against, or breaking specific rules. Please think before posting. If you feel induced to ask if something fits within the rules, it almost certainly doesn't.
ElkGroveDan
OK putting on the Admin hat here to elaborate on Doug's statement.

While we are all excited about the success of the unmanned Kepler Spacecraft and its results, this discussion is beginning to drift away from UMSF topic standards. We are going to try to keep this topic open since everyone is fascinated by the results. But let's keep this focused on the SCIENCE and ENGINEERING involved in Kepler's mission. We can discuss the results, the methodology, the equipment, the anticipated duration of the mission and potential duration of it's instrumentation etc. But let's get away from anything remotely related to astrobiology and all that it entails. Topics like 'habitable zones' are interesting but not really part of the larger UMSF goals and purpose, and there are plenty of other places to discuss those topics quite freely on the Internet.

Two admins in times zones on opposite sides of the globe are going to keep a very close watch on this and if we aren't able to keep it within the UMSF spirit and guidelines it will be closed without further comment or discussion.
SFJCody
QUOTE (ngunn @ Feb 4 2011, 08:55 AM) *
Me too. Where's it coming from? What do they want?


I think there are some holdouts who have yet to be convinced of the reality of any of the exoplanets discovered to date.
marsophile
It should be remembered that the results so far are still skewed towards short-period orbits. Since it will take several more years to confirm long-period orbits, any statistical conclusions need to be taken with a grain of salt. (Unless there is some reasonable way to extrapolate to long-period orbits.)
Eric Goldstein
QUOTE (ngunn @ Feb 3 2011, 04:55 PM) *
Me too. Where's it coming from? What do they want?


I'm generally too intimidated to post to this heavily moderated high quality forum, but since you asked....

My only reason for discontent is that the Kepler project (among other such projects) is withholding exciting data paid for with my taxes. Some people ask "what's the sense of urgency?" and to that I can only answer that we never know how limited our time is. I understand why proprietary periods are useful incentives, but I reject the notion that this is the very best way to get things done. I also reject the idea that the data must be withheld lest it be misreported -- science is a process, not a set of results, and can be reported as such.

SETI comments deleted. Please see Forum Guidelines section 1.3

I'm not bitter or angry - on the contrary, other than this one criticism, I'm absolutely thrilled with Kepler, and I am otherwise extremely supportive of the Kepler team.
Syrinx
All of Kepler's data is out and freely available now. Let's just leave it at that and be happy. Let's go out and get smashed this (Friday) night toasting to Kepler's tremendous (and continued!) success.
Juramike
QUOTE (SFJCody @ Feb 3 2011, 10:29 PM) *
I think there are some holdouts who have yet to be convinced of the reality of any of the exoplanets discovered to date.


It'd be neat to have independent types of evidence before describing world in a class such as "Super Earth" (odd that no-one seems to use "Super-Venus" or "Super-Mars"). A light curve is good enough to tell you there might be a world with a certain size and orbit, but a light curve + spectral information will be necessary to really determine if it is "Earth-like" (or Venus-like or Mars-like).

IIRC, the atmospheric CO2 content and planet size should be enough to indicate whether Earth-like plate tectonics are active or not. So if it is a big terrestrial and has a large CO2 content, it is likely a Super-Venus with little chance for smooth tectonics (and CO2 sequestering).

I also think it might be a good idea to refer to the worlds based on the triple-point zones of likely planetary volatiles: SiO2, H2O, CO2, CH4, and N2. That would give a much better idea of potential "Earth-like" geology processes like weather, rain, rivers, and oceans, and glaciers that would give a familiar landscape. (Is Titan Earth-like? Earth uses H2O the way Titan uses CH4 and the landscapes are eerily similar. If an exoplanet was big enough and cold enough to have liquid CO2, could we consider it "Earth-like"? Would a tectonically dead H2O ocean-planet be "Earth-like"?)

So while the Kepler discoveries are an awesome first step, there's one more step (and spacecraft mission) to go before I'd be comfortable describing it as a place I can compare/contrast with our own solar system's planets.
Mongo
One fact that has been noted by various people is that of those planet candidates in the "super-Earth" diameter range that have had their masses determined spectroscopically, they tend to be a lot less dense than expected for terrestrial planets of that diameter (so far). They basically look like Earths with a thick Neptune-like atmosphere. This immediately raises the possibility that there are a lot fewer terrestrial planets in the Earth size range than would be otherwise expected. Given the fact that all three roughly Earth-sized terrestrial planets in the Solar system have been plausibly suggested to have undergone a giant impact event shortly after they formed (Earth -- Lunar formation impact, Venus -- rotation-altering impact, Mercury -- mantle-stripping impact), it seems quite possible to me that such a giant impact event is necessary to strip away the thick primordial atmosphere (and water?), leaving an Earth-sized terrestrial planet. Mars might have been too small to accrete a dense Neptune-type atmosphere, or perhaps the hypothesized Borealis basin impact might have been large enough to do the job.

Is it possible that such giant impacts are relatively rare, and that the Solar system is unusual in that all (or almost all) its inner mini-Neptunes have had their Neptune-like atmospheres stripped away by their respective giant impacts? Perhaps due to the gravitational influence of the early Jupiter? This would certainly help explain the Fermi paradox, if true.
Drkskywxlt
Mongo...I think you're forgetting about orbital migration. Many of these sub-Neptunes or large super-Earths "should" have formed farther out in their systems...beyond the ice line. So they should have large volatile inventories, whether that's all water or water plus ammonia, methane, etc... Some studies showing terrestrial planet formation scenarios with a jovian planet out at 4-5AU show a wide variety of final water inventories...from "water worlds" to places as dry as the Moon, with all variety of masses. Jonathan Lunine was a co-author on one study, but I forget the 1st author.

Juramike...I'm not sure there would be much difference in the CO2 lines of Earth, Mars or Venus as seen from astronomical differences. I actually saw a talk by Victoria Meadows today (Virtual Planet Laboratory at UWash) and she said it would be challenging to infer surface pressures of extrasolar planets (not impossible though). You'd see the CO2 pop out very well but I don't think you could tell if it was many bars worth or just a few ppm. At least not initially. Whatever spectrum pops out from the first such mission, we'll need models to constrain the parameter space of planetary environments that could produce such a spectra.
Greg Hullender
Has anyone devised a way to look for the spectrum of water on a world that small? I know Kepler itself can't do it, but can anything else? Two really big discoveries would be a) another world with water oceans and cool.gif another world with an oxygen atmosphere. Heck, even ammonia oceans would be interesting.

--Greg
antipode
Another thing about all these large numbers being tossed around is that Kepler is only a 1m class telescope, so the Kepler volume is going to miss a potentially huge number of exoplanet transit candidates from stars at the bottom of the main sequence - stars that might otherwise be excellent transit candidates but that are simply too faint even though they might be fairly close by. That's going to take slow and painstaking RV and transit work by 8-10m class telescopes.

Note that I'm not buying into the arguments about potential habitability of M dwarf planets - that's for somewhere else...

P
Greg Hullender
Maybe not as much as you think. Check this out for some clever ideas to use ground-based telescopes of 2M and under to seek exoplanets around late-M dwarves.

QUOTE
There are compelling reasons to search for planets around late M-dwarfs: current planet surveys have not been able to target these optically-faint stars, and the reduced primary stellar mass improves planet detection sensitivity for new techniques which can target these stars. In this white paper we argue that one can employ aging 2 m telescopes for dedicated astrometric programs in the spirit of the RESTAR program to revitalize the US 2 m-class telescope population. Using a new low-cost adaptive optics system designed for 2 m-class telescopes we would be able to find planetary systems around late M-dwarfs with detection sensitivities approaching a few Earth masses.


http://www8.nationalacademies.org/astro201...lay.aspx?id=324

--Greg
SFJCody
QUOTE (Juramike @ Feb 5 2011, 05:11 AM) *
So while the Kepler discoveries are an awesome first step, there's one more step (and spacecraft mission) to go before I'd be comfortable describing it as a place I can compare/contrast with our own solar system's planets.



From what peter59 said, I got the sense that he didn't believe in any of the exoplanets, Kepler or non-Kepler, all the way back to 51 Peg.
Juramike
I think there is good evidence for exoplanets, and the Kepler team has done an outstanding job.

I'm just getting my feathers all ruffled about using the popular term "Earth-like" because that leads folks to imagine a planet with oceans, mountains, ice caps, rivers and the whole bit. And there is not yet evidence that those features exist. "Venus-like" is an equally valid assumption at this point.
ngunn
QUOTE (Juramike @ Feb 5 2011, 11:04 PM) *
I'm just getting my feathers all ruffled about using the popular term "Earth-like"


I've been feeling exactly the same. Why assume that the solar system provides a prototype for every possible type of planet? Even the few we have are all so different, it's difficult to see why exoplanets should resemble any one of our planets or even lie on a spectrum between any two. Size and distance from the star are not enough to go on.
nprev
Very true, guys. I suspect that we're ultimately going to find a surprising spectrum of possible topographies, climates, etc. that we really can't even suspect right now.

As usual, our provincialism will be revealed as precisely just that; the Universe isn't going to ever be what we want or expect it to be.
belleraphon1
Juramike, ngunn, nprev...

I concur. The variety and DIFERENCES from our little solar system is staggering.

Just look at the KEPLER11 system. Five planets with masses and radii between Earth and Neptune in our system, all orbiting inside Mercury's orbital radius. These are hot planets yet most of their densities are low. The sixth world inside Venus orbit.

What would the evening/morning twilight skies be like if Earth were transported to that G star.... would they be visible? Or all that mass hidden by the glare of Kelper 11?

No one predicted this just as no one really expected Peg51.

One thing we have learned since the unmanned spaceflight/telescope revolution of the mid-nineties is just how much we don't know.

Glorious....

p.s. Earth-like because, for one reason, KEPLER is funded by the public.... and the public could care less about the variety or worlds. They might prick their ears up about that Earth-like bit. And that is legitimate.
centsworth_II
Its all a matter of definitions. Mercury is a Terrestrial planet, but not an Earth-like planet if orbital zone is considered.

According to http://merriam-webstercollegiate.com/dicti...mp;t=1297015687 Terrestrial planets are those "belonging to the class of planets that are like the earth (as in density and silicate composition) <the terrestrial planets Mercury, Venus, and Mars>." So planets that may be more like Venus, Mars, or Mercury can still be considered "Earth-like".

Right now the limits of observation only allow a planet to be called "Earth-like" based on size/mass and orbital zone. Knowing if a planet actually more resembles Venus or Mars is out of the question. If a Kepler-like instrument looking back at our solar system from a distance of many light years could not determine if Venus is in the habitable zone, it could not be ruled out as an Earth-like planet on the basis of orbital zone.
Syrinx
Adding to what centsworth has said:

None of the Kepler-11 planets qualify as Earth-like. They're all too massive and too close to their sun. However, Kepler and ground-based follow-on observations will in time hopefully uncover "Earth-similar" planets based on mass, radius, and surface temperature. Going any further than Earth-similar ("Earth twin") will require a spectroscopy mission.

Way back before we had results to discuss, we developed our own guide for "Earth-like" halfway back in this very thread:

http://www.unmannedspaceflight.com/index.p...mp;#entry149307
ngunn
I have a question about Kepler 11. How edge-on is it? If there were planets further out would they fail to eclipse?

I think the duration of eclipses will have provided some inforrmation.

(Sorry, I don't have access to the paper. It's probably in there.)
Hungry4info
If there are further planets, it depends entirely on their orbital inclination whether or not they would eclipse.
Also for the planets at Kepler-11, we know their inclinations against the plane of the sky are within 2-3 degrees of each other.

Here's free access to the paper.
http://arxiv.org/abs/1102.0291
Greg Hullender
@ngunn Yeah, given the period, the estimated mass, and the estimated diameter of the star, the duration ought to give you the inclination. Without the period, onset time (first contact to second contact) can also give you the inclination, but at Kepler's cadence, that's hard to measure. Alternatively, given longer times to observe these things, one ought to be able to use the onset times AND the period to refine estimates for the other parameters.

They'll be mining this data for twenty years, I expect . . .

--Greg
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2014 Invision Power Services, Inc.