An Apodized Kepler Periodogram for Separating Planetary and Stellar Activity Signals

TL;DR

This paper introduces an apodized Keplerian model for analyzing radial velocity data, effectively distinguishing planetary signals from stellar activity by using Gaussian apodization and Bayesian methods, improving detection accuracy.

Contribution

The paper presents a novel apodized Keplerian model combined with Bayesian MCMC and a differential Lomb-Scargle periodogram to better separate planetary signals from stellar activity in RV data.

Findings

Achieved a sixfold reduction in stellar activity noise.

Successfully identified planetary candidates in simulated data.

Enhanced signal separation with an augmented moving average component.

Abstract

A new apodized Keplerian (AK) model is proposed for the analysis of precision radial velocity (RV) data to model both planetary and stellar activity (SA) induced RV signals. A symmetrical Gaussian apodization function with unknown width and center can distinguish planetary signals from SA signals on the basis of the span of the apodization window. The general model for $m$ apodized Keplerian signals includes a linear regression term between RV and the stellar activity diagnostic $\log(R'hk)$, as well as an extra Gaussian noise term with unknown standard deviation. The model parameters are explored using a Bayesian fusion MCMC code. A differential version of the Generalized Lomb-Scargle periodogram that employs a control diagnostic provides an additional way of distinguishing SA signals and helps guide the choice of new periods. Results are reported for a recent international RV blind challenge which included multiple state of the art simulated data sets supported by a variety of stellar activity diagnostics. In the current implementation, the AK method achieved a reduction in SA noise by a factor of approximately 6. Final parameter estimates for the planetary candidates are derived from fits that include AK signals to model the SA components and simple Keplerians to model the planetary candidates. Preliminary results are also reported for AK models augmented by a moving average component that allows for correlations in the residuals.