Skip to content
Planet 47

Characterizing and understanding the planet population around intermediate mass stars

Principal Investigators

  • Dr. Sabine Reffert
    Landessternwarte Heidelberg-Königstuhl, Heidelberg


So far, Doppler surveys of evolved stars have revealed many planets orbiting intermediate-mass (IM) stars (M>1.5 solar masses). Compared to the planets around less massive main-sequence stars, the population around IM stars exhibits prominent differences; most notably, close-in planets such as Hot and Warm Jupiters seem to be very rare. The goal of this project is to develop a scenario in which the origin of the observed differences can be understood.
For over 12 years, our group has carried out an extensive Doppler survey for planets orbiting giant stars at Lick observatory, which will soon be continued with a newly built spectrograph for the Waltz Telescope on the Königstuhl in Heidelberg. This will help to further characterize the planet population around IM stars by uncovering planets with periods of the order of decades, for which we have excellent candidates in our sample.
Additionally, we will place the known properties of the planet population around IM stars on a more secure statistical footing by combining our survey results with those of the EXPRESS and PPPS giant star surveys. The joint analysis will derive updated planet occurrence rates and their dependence on host star properties such as mass and metallicity, while taking the detection probability functions for each target star into account.
As most of the known IM planet hosts are evolved stars on the giant branch, the stellar evolution could have had an important impact on the architecture of the planetary systems. We will carry out simulations on how the planet orbits change due to stellar mass loss and tidal effects during the giant stages of stellar evolution. Running our simulations backwards in time allows us to determine the orbital configuration as it used to be during the main-sequence phase, starting out from the currently observed evolved population. So far, it seems that stellar evolution alone cannot explain all the differences. However, there are still many unknowns in the treatment of tides for giant branch stars so we aim at exploring a wider range of input parameters and mechanisms to ultimately reconcile the discrepancy between the observed populations.


Vera Wolthoff


Hi, my name is Vera and I’m a PhD student at the Landessternwarte Königstuhl in Heidelberg and a member of the project “Characterizing and Understanding the Planet Population around Intermediate-Mass Stars”. Most planets found so far around IM stars have been detected via the radial velocity method. However, IM stars are difficult targets for spectroscopic planet detection during the most part of their lives: they are much hotter than their lower-mass counterparts so they only show few absorption lines in their spectra, and they rotate faster leading to rotational broadening of the lines. Both reduces the achievable RV precision. But when IM stars run out of fuel in their cores and evolve into red giants, they cool down and rotate slower becoming feasible targets for RV planet searches. So dealing with planets around IM stars usually means dealing with planets around evolved stars – which also means that observed differences to the planet population around lower-mass stars can be caused by either the higher stellar mass or the advanced evolutionary stage, or both.

On the one hand, my research aims at statistically characterizing the planet population around IM stars. For this, we have joined forces with two other giant star surveys – the EXoPlanets aRound Evolved StarS (EXPRESS) survey and the Pan-Pacific Planet Search (PPPS) – to complement the observations from our own Lick giant star survey, which gives us an overall sample of more than 600 stars. However, we have to deal with inhomogeneous observational coverage and varying radial velocity precision among the three surveys. So for a statistical assessment, we need to quantify which kind of planetary signals would be detectable given the observational history and stellar noise of the target stars. This is done by inserting simulated planetary signals with different periods into the RV data and increasing the amplitude of the signals until they become detectable. This gives us a detection probability function in period and planet mass space for each target star which we can then use to compute planet occurrence rates and determine how they depend on stellar mass and metallicity.

Comparing the planet population around evolved IM stars to that of lower-mass main-sequence stars, the most notable difference is that close-in giant planets (Hot and Warm Jupiters with P<90days) are very rare around our target stars. Hence, the second goal of my research is to understand the reason for these differences. The stellar evolution on the giant branch can have different effects on the orbiting planets: on the one hand, the star loses mass which can cause planetary orbits to expand; on the other hand, the stellar envelope becomes deformable and the planets can raise tides on the star whose torque usually leads to shrinking of the orbits, and may ultimately result in engulfment of the planet. Our simulations of these effects can be run backwards in time to recover the planetary orbits as they were before the star started evolving into a red giant. So far, it seems that for stars more massive than 2 solar masses, the simulations cannot explain the observed differences in period distribution. However, this conclusion has to be taken with a grain of salt as only equilibrium tides were considered because the handling of the so-called dynamical tides (interactions between tidal modes and stellar oscillations) for giant stars is very uncertain. This is why in the next step, we want to turn the question around: Given the observed planet distribution, how strong would the tides need to be to explain the differences to the unevolved population?

Invited Guests


To top