Kepler-37 : A Sub-Mercury, a Sub-Earth and a Super-Earth

Newly discovered planet is tiniest one yet

http://www.cbc.ca/news/technology/story/2013/02/20/science-smallest-planet-discovered-kepler-37b.html

NASA Page.
http://www.nasa.gov/mission_pages/kepler/news/kepler-37b.html

How to measure a mass of the planet b ?

It's probably not possible with current technology. And probably not for the foreseeable future.

Any ttv between b and c?

For planets this low in mass and this widely separated, I would not expect there to be any detectable TTVs. Hopefully the paper will mention this when it comes out.

Paper on nature.
http://www.nature.com/nature/journal/vaop/ncurrent/full/nature11914.html

Very beautiful system, spanning the whole range of telluric planets from Moon size to Super-Earth (assuming that Kepler 37-d will turn out to be a mainly rocky planet a là 55 Cancri e).

Kepler 37 d is rather water world but radius of the star is somewhat too small for G star and t eff 5520K.Mass and radius typical for K 0-2V

An [M/H] of -0.32 dex = 48% is enough to heat up the atmosphere into G. 5417 K is about G8.5, which makes Kepler-37 markedly similar to Tau Ceti (though a spot more metal-rich, and therefore more massive)

So they've found a planet smaller than Ganymede.

Hmmmm.

Now if only we can get a transiting jovian planet at a large enough distance around a sufficiently quiet star...

(Though I guess figuring out how to fold the data properly to find an exomoon would be tricky...)

I have computed a transit probability of 1.72%, for planet d, assuming a circular orbit.

Taking into account less than 500 G stars of 10.5 mag or brighter, in the Kepler field, only with this star we have at least 12.4% of G stars with inner planetary systems of this type.

Lazarus: to find Ganymede-like planets you need a lot of transit events in order to reach enough SNR. Jovian planets at large enough distance haven' t enough events to detect that tiny transits.

And... what about candidate .04 at 50 day period?

Kepler-37 is a very interesting, bright and nearby star.... It will probably be study deeper in the future.

Planet b is hardly larger than Io -extreme performance but to find it around jovian at 1AU- more , we need some transits and a distance from moon to planet the larger the better.Rings could be problem also to recognise between them

@pochimax: well according to Barnes & O'Brien (2002), Gliese 86 b (orbital period 15.8 days, versus 13.4 days for Kepler-37b) ought to be able to retain a satellite with about 0.1 Earth masses, which is already more massive than Ganymede.

Of course the transits of the satellite won't occur at regular intervals... and there are other issues such as the occurrence of giant planets at suitable orbital periods, the survival of satellite systems during planet migration, etc.

pochimax wrote:

And... what about candidate .04 at 50 day period?

According to the supplementary information, the significance of .04 has decreased:

« …Any ttv between b and c?… »
(tommi59)


http://www.readcube.com/articles/10.1038%2Fnature11914

« …It was not possible to confirm using radial velocities or transit
timing variations that the three candidate planets are substellar
bodies orbiting Kepler-37. We therefore explored possible astrophys-
ical scenarios (blends) that can mimic a planet transit across the disc of
Kepler-37 using the BLENDER procedure. BLENDER analyses
attempt to show that a blend scenario is much less likely than a planet
interpretation by creating a wide array of synthetic light curves of
various blend scenarios and comparing the goodness of fit to the syn-
thetic light curves with that of the true planet model. Inconsistent fits
are rejected, as are fits to scenarios that are ruled out by additional
information such as high-contrast imaging, Kepler Input Catalog colours
and high-resolution spectroscopic observations. Details of
the supporting data used by BLENDER are provided in sections 2–5
of Supplementary Information. The final blend probability is com-
pared with the expected frequency of occurrence of real planets
(known as the planet prior) to calculate the final probability that the
candidate is a true planet. Our planet priors are calculated by looking
at the frequency of planets with a size within the 3s uncertainty range
of the measured planet radii. We then apply a correction to the
planet prior to account for false positives and incompleteness in the
planet catalogue… »

Paper on arXiv :
http://arxiv.org/abs/1305.5587

Looks like a fourth planet at Kepler-37 (KOI-245). A 0.267 Earth-radius planet (pebble?) in a 51-day orbit.
http://arxiv.org/abs/1310.7942

Yeah that crossed my mind. Do you think maybe the significance of the fourth signal has increased with the addition of more data?

The NASA Exoplanet Archive has added Kepler-37 e.
http://exoplanetarchive.ipac.caltech.edu./cgi-bin/DisplayOverview/nph-DisplayOverview?objname=kepler+37+e&type=CONFIRMED_PLANET

With this paper as the reference
http://adsabs.harvard.edu/abs/2014ApJ...787...80H

Well now. This seems like a bad idea, considering the kepler team believed the candidate signal was noise. It's still listed as a false positive in the KOI archive, anyway. Hopefully that'll stick.

Anyway, did you notice that the paper updated in April? What's with that?

Some added planets today have very high massess-collisional stripping? http://exoplanetarchive.ipac.caltech.edu./docs/exonews_archive.html#29May2014
There is some problems with 4th signal 51 days in kepler 37 system due much less transits than planet b