Sunday, August 2, 2015

All sky map of exoplanet host stars (

Today I'd like to show something not directly connected to the last couple of posts. This is an all sky map of exoplanet host stars. Different colors/symbols indicate the method used to detect the planet. The data was taken from Actually, I tried to plot all planetary systems know today, which according to should be 1228, but for some reason I am missing three in the transit method.

The underlying magenta colored points are about one million stars from the Tycho catalog (ESA Hipparcos satellite). The map is an aitoff projection of the sky in galactic coordinates and the Milky Way lies at the equator; this is why most of the stars are located there.

Most of the planets are detected with the transit method. However, the majority is located in the small area crowded with black dots - the Kepler field of view. When we talk about statistics of transiting exoplanets we are actually talking about a small part in the sky and not the entire sky. Astrophysicists just assume that it should be the same everywhere. The other black dots come from different surveys, e.g., CoRoT or WASP.

The picture is different when looking at planets detected with the radial velocity (RV) method. They seem to be more uniformly distributed. It's the second most successful technique with 454 stars having planets around them.

At least according to, so far there is only one astrometry planet and it is very close to the Kepler FOV - although it's not in it. The star is HD 176051 b. So far it seems to be too difficult to detect planets with this method, but it is expected that with GAIA there will be many planets coming from this technique.

Imaging seems pretty much located to certain areas, avoiding the plane of the Milky Way as much as possible. This is probably a good idea since background stars that coincidentally stand close to the potential host stars make it more difficult to find planets - or might even be misinterpreted as bodies belonging to the system.

All microlensing planets come from a limited area in the direction of the center of the Milky Way, which in this map is at the edge. This illustrates nicely that microlensing really samples a quite different population of stars. All other techniques try to go away from the galactic center (and even the galactic plane) and only find exoplanets rather close to the Sun. The most distant microlensing planet, however, is about 25000 light years away - virtually in the center of the Milky Way.

The transit timing variation (TTV) planets are all in the Kepler FOV. For this method one needs good light curves and a long cadence to cover many transits, and this is what Kepler does best. According to there are only four TTV planets - which I do not think is true; there have to be much more. says it's more like 60 (including some pulsar planets), which I believe is closer to the real number.

Finally, we have the planets around pulsars. There are a few in the Kepler FOV, but otherwise I do not really see a system there. I think it is nice to notice that PSR 1257 12 b is the star with the first exoplanet detection in the year 1992 - and it actually is a three-planet system! I think is has to be one of the two triangles in the upper right. Usually, people cite 51 Peg b as the first exoplanet - it was the first around a solar-type star. I guess pulsars are just too different ...