All Features Great and Small: Distinguishing the effects of specific magnetically active features on radial-velocity exoplanet detections

Abstract

State of the art radial velocity (RV) exoplanet searches are limited by the effects of stellar magnetic activity. Magnetically active features, such as spots, plage, and network regions, each contribute to the observed RV shift through a variety of mechanisms, including via the suppression of convective blueshift and by creating rotational imbalance due to brightness inhomogeneities. However, the extent to which these RV contributions depend on the specific properties of individual active regions remains unknown. In this work, we investigate the effects of active region size on activity-driven RV variations, and develop tools for modelling these RV variations on Sun-like stars. We analyze solar observations acquired over Carrington Cycle 24 to test models of stellar magnetic activity and the resulting RV variations of Sun-like stars: we compare direct measurements of solar plage, spots, and network using the Solar Dynamics Observatory (SDO) to measurements of the solar RV and S-index from the solar telescope at the High Accuracy Radial velocity Planet Searcher for the Northern hemisphere (HARPS-N), solar photometry from the Solar Radiation and Climate Experiment (SORCE), and variations in the Sun's acoustic oscillation frequencies from the Birmingham Solar-Oscillations Network (BiSON). By comparing estimates of the contributions of the suppression of convective blueshift and the rotational flux imbalance derived from SDO images to the HARPS-N solar RVs, we find that that magnetic active regions smaller than 60 Mm^2 do not significantly suppress convective blueshift. Differentiating the relative coverage, or filling factors, of these large plage regions from small network regions is thus necessary to differentiate between activity-driven RV signatures and Doppler shifts due to planetary orbits.

We then investigate several methods for extracting the relative coverage, or filling factors, of spots, plage, and network features. We demonstrate that variations in the solar p-mode frequency are highly sensitive to the presence of large plage regions, and are significantly less sensitive to smaller network regions, making the p-mode frequency a possible — though observsationally intensive — avenue for differentiating the contributions of different classes of active regions. We then develop a technique to estimate feature-specific magnetic filling factors on stellar targets using only spectroscopic and photometric observations. Linear and machine learning implementations of this technique both yield filling factor estimates that are highly correlated with the observed values. Modeling the solar RVs using these filling factors reproduces the expected contributions of the suppression of convective blueshift and rotational imbalance due to brightness inhomogeneities, providing an avenue for estimating these RV contributions on well-observed, Sun-like stars.