Nice update on Spaceflight Now, with some new (to me) information:

http://www.spaceflightnow.com/news/n1007/28kepler/



I think Earth-sized is a very fine term.



I suppose this shouldn't surprise me, but it does. I'd have expected them to have produced this software before they launched Kepler. I realize they had more CCD problems than originally expected, but still.



That seems like an unbelievable amount of time for a data-conversion operation. I suppose that includes heavy-duty things like wavelet transforms, but even so. Also, this makes it sound like they are gathering data about ten times as fast as they can convert it. Maybe they should write it so it can run on a home PC and then invite ordinary enthusiasts to help. Yeah, it'd take time to write the software to do that too, but if they're looking at 40 years to convert the data, that might not seem so bad.

Or maybe I'm just entirely misunderstanding it. :-)

--Greg

Given that Gravity Probe B is still tinkering with their software, it's quite alright for Kepler to be debugging...

I think that a Kepler@Home project, if technically feasible, would be a fantastic idea. Not only would it make a ton of computation power available, it would give the scientifically literate public a chance to take some kind of ownership of exoplanet discovery. And that could be a very powerful thing, boding well for future missions.

I would be surprised if that was limited by raw CPU time, and also surprised if it was just data conversion rather than analysis. My guess is that the pipeline is still being developed, and the four months isn't how long it will take in general, but how long it has taken to date. Rolling out a "kepler@home" would before the process is nailed down would involve significant extra development effort.

The text you quoted does suggest your interpretation, but notice that it's the reporters words and not a direct quote from the mission team.

Another opportunity to hear Bill Borucki speak will be coming up August 5-8 in Dayton, Ohio, at the Mars Society Convention. Borucki will deliver the plenary address. For more information, see http://www.marssociety.org/portal/kepler-m...iety-convention.

Well, I assume there will be a Kepler II mission at some point. It would be a great idea for someone with copious free time to take the raw Kepler data and develop a BOINC workflow, calibrating etc along the way, so that the application can double-check the final results, and also prepare for follow on transit searches. Hopefully there will be a fleet of Kepler follow-ons to search the rest of the sky for transits. Perhaps there is no need though. None of the Kepler stars is really close enough for imaging planets even with TPF, right?

IIRC, the high-end version of TPF could potentially resolve large-scale surface features on planets out to something like 20 light-years. (Gotta admit that still sounds implausible to me, though.) Kepler's target stars are all well over a thousand light-years away.

Google Galaxy!

You can watch the light curve over the course of a planet's rotation period to resolve surface features longitudinally.

If the planet rotates several times per orbit, you can watch the light curve over the course of the planet's orbital period to resolve surface features latitudinally.

It's kind of how they build the map of Pluto, an indirect method.

Yes, I recall the Planet Imager (PI) that had been proposed as a network of TPF style interferometers that could image some nearby Earth-sized planets with 25 pixel resolution.

I think it's all but inevitable that we will end up probing other planetary systems to the limit of the physically possible eventually, even if it's not within the next 25 years. Given the near impossibility of interstellar travel, the only means of exploring our galactic neighborhood that will be available to us for decades, and very likely for centuries, will be telescopic missions, be they in Earth's orbit, at L2, in the outer reaches of the solar system, beyond the zodiacal dust cloud, and even to the focal point of the Sun's gravitational lens at a distance of 550AU.

And yes, there will be fleets of interferometric telescopes one day, because unless we quickly reach the theoretical viewing limits, it will continue to be our only means of imaging and gathering more information about extrasolar planets.

So Google Galaxy is right. We're almost certainly going to have a catalog of millions of planets one day, even if we can only take images sharp enough to create maps of the surface features of the nearest few thousand or so.

Good times...

http://www.seti.org/colloquium

Almost six weeks since the last manager's update. Wonder if that means they are gearing up for a big announcement in September? Waiting for another batch of papers to finish peer review? I hope so. But then again, maybe they just forgot about the update thing.

Here it is!

http://www.kepler.arc.nasa.gov/news/mmu/in...s&NewsID=55

Says their propellant use is lower than expected, so they could go 10 more years, if they got funding. That'd be nice. Confirms the CCD that failed in January is still dead, but says that has only a small impact. They do have an issue that their data transmission rate drops over time as the spacecraft gets further and further away.

They talk at some length about how the Followup Observation Program (FOP) helps them vet the hundreds of planet candidates to eliminate false positives.

--Greg

Have they considered looking at a different field-of-view during the extended mission? Or would the goal be just to find longer period planets in the current FOV near Cygnus?

Yes, and yes.

Indeed it seems that the same FOV would yield decreasing results as the years pile on. But remember, Kepler does not function in isolation. Before Kepler launched, the Kepler ground team studied thousands of stars in the current FOV. The study enabled precise categorization of parent stars and therefore more accurate and precise results for their planets. The team would need to do the same study for any prospective FOV, and it's unlikely they'll find the time or the funding.

But, 10 years is a long time. Plans change.

I just got a Tweet from Kepler saying there's a press conference Aug. 26 at 10am PDT to discuss "an intriguing star system."

-- edit --

Link:

http://www.nasa.gov/centers/ames/news/rele...10/10-72AR.html

Could this be considered as intriguing too...?

http://www.eso.org/public/news/eso1035/

Fascinating news, and a great achievement to be sure, but personally I just can't bring myself to think of an object with a "year" of just over 1 day as a planet. Something moving that closely and quickly around a star isn't a planet. It's a moth.

Good points. If they got ended with a total mission time of ~10 years, that should allow transiting planets with periods up to 5 years detected. But, with such long periods, you're looking at a REALLY narrow window of orbital inclinations to allow transits from our POV. I hope they'll pick a new FOV, but I imagine they'll stay on this one and just refine parameters on the planets they do find as well as look for longer period planets.

Kepler News conference this Thursday ... intriguing star system ... yikes! ... cant wait!
<speculation>
part of the system has a (large) planet (with potential moons) within the calculated habitable zone...
</speculation>

Well, the other four potentials sure fit that category then.

That would certainly be a find! But, I doubt it. I looked for moons in the Kepler data for the first 5 published planets (and didn't find any). What I did find out is that it will be a pretty amazing achievement to find a moon by transit-timing variation or transit-duration variations. The reasons are :
1. To get a big TTV or TDV, you want a small planet/big moon combination close to the star. In a close orbit, you get more precise transit times, which can let you tease out those TTVs or TDVs a little better. But, it's very unlikely for a large moon to be stable over a long period of time in a Hot Jupiter type orbit...ignoring the probable difficulty of giant planet migration while keeping its retinue of moons.
2. So, with planets farther out (closer to their original orbits), you're more likely to have stable very large (Mars or Earth-sized) moons, but it's going to be much harder to find transits on those planets and even harder to get precise TTV/TDV data.
David Kipping has a host of papers on the topic of finding moons with Kepler data.

I suspect their announcement will be a multiple planet (3 or 4 planet) system. Perhaps one will be in the habitable zone, but I bet one will be ~1-1.5 Earth masses. I bet their a bit bummed that the Europeans scooped them!

I think that after 6 years of looking at the current FOV, they should move on to another. 6 years gives enough time for 3-9 transits of semi-earths (say, 8 months to 2 years orbit time), which should give enough information to do significant statistics on. Also, once found, followup studies can be conducted from Earth.

Edit: And for the next FOV, choose one with a different galactic population, ie different metallicity and average age. That would help inform models of system formation.

Oh, absolutely they do, wasn't belittling that at all. And I'm really looking forward to the Kepler announcement! But a "year" that's a day long? That's just... silly.

I understand your argument, Richard, but at the same time it might be wise to constrain the mission's objective appropriately. Is it really to find possible Earth-analogs, or is it to get some idea in a statistical sense about the overall nature & frequency of occurrence of planetary systems in general around Sunlike stars?

From a scientific viewpoint, I'd argue that that latter objective would be more productive at this stage of the game. Until Kepler, we've generally only been able to see large planets in tight orbits around other stars, but we knew already that's not the only possibility since we don't live in a Solar System like that. The sad fact of the matter is that if Kepler finds an Earth-sized planet (or many!) dead smack in the middle of a given star's Goldilocks zone then what I just wrote is literally all we're gonna know about it for quite awhile. The discovery of these bodies hopefully would spike enough interest to fly the best TPF we can as a follow-on to Kepler, and that's when the real science on that very interesting niche of planetology will happen.

Meanwhile, for as long as Kepler survives, I think that X^nMs should be spent gathering as much data as possible about the distribution of planets & the nature of solar systems. In this case, that would mean staring at this same patch to map as many solar systems as completely we can.

Woooooooooh!!! Every time I think about that I genuinely get shivers running up and down my spine. Just imagine THAT day, when we're all sat here, at our computers, waiting for the press conference from the TPF Team. There are rumours that they've found a true Earth-like planet, and have images, but NASA has stayed silent on the speculation, and amazingly no info has leaked out. Then the conference begins and an image flashes up on the screen... a tiny ball, just a couple of dozen pixels wide, but it's blue, and white...

That's why I love Kepler so much. With Kepler we're all standing on the shore of Sagan's cosmic ocean, looking out to sea. We've known there must be other islands out there, other beaches, but we've had no proof. Until now.

Amazing times we live in. Amazing times...

Well, not any more silly than a day that's longer than a year!

I dunno... I frequently have days at work that seem longer than a year...

That would require something a bit more advanced than TPF.

http://www.daviddarling.info/encyclopedia/...anetImager.html

Mmmm water cloud Jovians...

- Matthew Holman, associate director, Theoretical Astrophysics Division, Harvard-Smithsonian Center for Astrophysics, Cambridge, Mass.
- Alycia Weinberger, astronomer, Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington

Hum, what does the inclusion of these two tell us, if anything?

- Edit -

Both have published in circumstellar debris disks...

I recall that some versions of TPF or TPF-like exoplanet missions would be able to produce "images" a few pixels wide. Even that would possibly be able to determine the rough locations of continents, time the planet's rotation, and maybe even detect sun-glint off its oceans. On the sun-glint, an article was just posted on Arxiv yesterday saying that JWST would be able to detect sunglint on exoplanet oceans IF it had an external occulter flying nearby. http://fr.arxiv.org/abs/1008.3864

As long as we're taking bets, I'm thinking it's just about time for them to announce the first "hot Earth". Something Earth-sized or smaller with dozens or even hundreds of transits.

As for the extended mission, it does seem Kepler suffers from the law of diminishing returns in two ways. First, as already mentioned, planets in longer orbits are much less likely to transit at all. Second, as far as reducing the uncertainty of its other results, that only goes down with the square root of the number of observations. So the difference between ten years vs. four years only knocks about 35% off your error bars. On the other hand, maybe they might find indirect evidence for other planets, so that'd be an argument for sticking with the same spot.

--Greg

Of course, it would. I was following Nick's train of thought that Kepler success might lead to follow on missions that were more advanced than originally envisaged. So I believe that a post-Kepler TPF would be a lot more powerful than the TPF originally proposed.

There was a follow on mission to the tpf loosely specified that I believe had a 5x5 pixel rendition. It was called the Planet Imager and was an array of TPF's in a formation.

nprev,
Everyone else already said it, but to reiterate: Once we've got 3 transits, minimum, of anything with <2yr orbit, I think it's time to move on to a new FOV. Remember, that doesn't mean Kepler won't have found anything with a 5 year orbit, but that most objects of (likely transit)*(interest to us) would have been found. Moving on to a different view would tell us more about solar system formation and how it depends on variables like star metallicity.

Besides, I don't see Kepler as finding every planet, but rather finding interesting planets worth spending time and $$$ on following up. An exo-Mars or Neptune orbiting in a 5 year orbit isn't likely to warrant followup.

Basically, I think we get better statistics on things like "how many earth analogs to expect per 100 cubic parsecs at Sun's distance from the galactic core" when we widen the net vs trolling (hah, that must be the first time in the history of the internets that word is actually used as originally defined) the same area.

Edit: Also, I don't think the Kepler stars are necessarily great prospects for a TPF followup anyways, since they are intentionally fairly far away, to keep them from being too bright in the Kepler CCDs.

What Kepler can really tell us is stuff like "Expect 1.3 semi-Earths for every 100 G stars you look at with metallicty X, 2.6 sE for every 100 K at with metallicity Y, etc." Basically, a 3D graph with star mass on the X axis, star metallicity on the Y axis, and sE probability on the Z. This could then direct a TPF "planet hunt" targeting the nearest stars most likely to host Earth-like planets.

Really, what I would do after the nominal mission is look at new FOVs every 2 years, to maximize that data set. And then look into solar sailing to keep going once the tanks are empty.

If this were CoRoT, I'd agree.

But finding the most planets is not the objective of Kepler. Kepler is trying to make a census of planets at a range of orbital periods. The range of these orbital periods probed will depend on how long it stares at the same area of the sky. The Kepler candidates are not going to be very accessible to anything other than noisy radial velocity to constrain their mass. That's it. There is little to be gained by switching fields of view... certainly not enough to make up for the loss of probing the planet population at greater distances for the star.

hendric,

Absolutely correct. Direct planet imaging will be concentrated on the closest planetary systems. The 5x5 pixel view I alluded to with the "planet imager" applied to only very close in systems. I agree completely, Kepler is all about statistical characterization of stars and planetary systems. With that information, the design of follow on missions can then be done (much more) intelligently. I think we will want two+ missions for close in planetary systems. The first mission targeted to identify at least a handful of the closest earth sized planets in habitable zones. And the second (and other) missions targeted to muster the highest resolution multi-spectrum views of those worlds and parent star systems. I think in about 20-30 years we will finally see some images of alien "earths." Even before that we will get substantial information back on them through the ingeniuty of astrophysicists (e.g. statistical study of reflected light off of rotating and orbiting exo-planets).

I'm confused is this televised?

nope, if there are any additional questions regarding this release they will gladly be answered at Kepler's FB official page, drop by!
http://www.facebook.com/NASAsKeplerMission?ref=ts

Not televised but you can listen to streaming audio at http://www.nasa.gov/news/media/newsaudio/index.html

"Link will appear..."

FWIW, there's this bit from the Kepler web site:

The Kepler Mission, NASA Discovery mission #10, is specifically designed to survey our region of the Milky Way galaxy to discover hundreds of Earth-size planets in or near the habitable zone and determine how many of the billions of stars in our galaxy have such planets.

Based on this, it would seem a new FOV would be more in keeping with the mission as described; but really I don't know enough to have an opinion either way.

Link at http://www.nasa.gov/news/media/newsaudio/index.html is now up.

Supporting documentation is available at http://www.nasa.gov/mission_pages/kepler/m...ia_telecon.html

And off it goes!

Andy

Hmmm..."Kepler 9". (Sounds a bit Star Trek to me!)

Two transiting saturnoids, orbiting a "sun-like star" in a 2:1 resonance of 38 days and 19 days. Orbital parameters change slightly over time, suggesting a third planet candidate. This is a possible hot-Earth of 1.5 Earth-diameters much closer in.

The news in brief: "Demonstration of transit-timing alteration allowing for otherwise unseen planets to have their orbits calculated - allowing for 'discovery' of planets that may lie in the habitable zone of stars even if they're not seen transiting."

umph.

First detection of multiple planets in one system using transits. (Is that right?)

Anyway, http://kepler.nasa.gov/Mission/discoveries/

Count of confirmed planets is now 7.

The Kepler Planet Counter on the website should be updated soon...it still stands at 005.

There is the link to very interesting paper about recent discovery.

It's been updated; 007 now.