How does stellar convection impact the detection of small planets at high radial velocity precision?

One of the most successful methods of detecting planets outside of our solar system (exoplanets) is the radial velocity technique. It utilizes the fact that, for a planet hosting star and its companion, the planet is not simply orbiting the star but both are orbiting around a common center of mass. This periodic motion can be measured and allows inferring the presence of a planet, otherwise invisible to the observer. This is done by measuring the Doppler shift of signatures of specific elements in the stellar spectrum, which is its light separated by wavelength, or color, like a rainbow in the sky. This effect is similar to an ambulance’s sound changing pitch up and down as it moves towards or away from a listener. Light undergoes a similar shift with objects moving towards an observer looking blue shifted, i.e. towards shorter wavelengths, and objects moving away red shifted, i.e. towards longer wavelengths. By knowing where a specific feature is supposed to be located, the velocity along the line of sight can be determined.

Due to its much higher mass, the motion of a star however is many orders of magnitude smaller than that of the planet. Even the most massive planets known induce variations of only a few hundred meters per second on an object hundreds of thousands of kilometers large, with an earth-sun like system only moving at a level around 9 centimeters per second. While this used to pose a significant problem in the past decade, the requirements for instrumental stability are immense to reach this level of precision, present day instrumentation is now up to the task. Unfortunately this has shifted the limiting factor from earth-bound instrumentation that can be mitigated by continuous innovation to the observed stars themselves.

A stellar surface, when observed up close, looks much like a boiling pot of water. Hot, and in the case of stellar material very bright, material rises up in large granules to cool at the surface and then flow sideways into narrow lanes where, now much darker, it can sink back down again to rejoin the cycle. This motion, called convection, asymmetric in surface area and brightness and ever changing in time, is one major contributing factor towards variability intrinsic to stars since the material rising at any point appears blueshifted and the sinking one redshifted. This unavoidable variability can easily introduce uncertainties of several meters per second, completely hiding the signal of earth-like planets.

Studying these phenomena is therefore of utmost importance to enable us to detect earth-like planets. Only if one understands how this variability comes to be, how it develops in time, how it changes from star to star and even from spectral feature to spectral feature can it be possible to disentangle intrinsic stellar effects from planet induced motions. Reaching that point is the goal of this project. By combining theoretical predictions with empirical data, a combined model is to be created, allowing to subtract stellar convection effects from observed data to recover the planetary signal.