DIAMONDS: A new Bayesian nested sampling tool

TL;DR

Diamonds is a new Bayesian nested sampling tool designed for efficient high-dimensional and multi-modal parameter estimation, demonstrated on stellar oscillation data, with a novel multimodal peak bagging approach that simplifies analysis.

Contribution

The paper introduces Diamonds, a Bayesian nested sampling code optimized for complex, high-dimensional, and multi-modal problems in asteroseismology, including a new multimodal peak bagging method.

Findings

Successfully analyzed solar-like oscillations in a challenging F-type star.

Effectively distinguished different stellar background components.

Reduced free parameters in peak bagging through multimodality approach.

Abstract

In the context of high-quality asteroseismic data provided by the NASA Kepler mission, we developed a new code, termed Diamonds (high-DImensional And multi-MOdal NesteD Sampling), for fast Bayesian parameter estimation and model comparison by means of the Nested Sampling Monte Carlo (NSMC) algorithm, an efficient and powerful method very suitable for high-dimensional problems (like the peak bagging analysis of solar-like oscillations) and multi-modal problems (i.e. problems that show multiple solutions). We applied the code to the peak bagging analysis of solar-like oscillations observed in a challenging F-type star. By means of Diamonds one is able to detect the different backgrounds in the power spectrum of the star (e.g. stellar granulation and faculae activity) and to understand whether one or two oscillation peaks can be identified or not. In addition, we demonstrate a novel approach to peak bagging based on multimodality, which is able to reduce significantly the number of free parameters involved in the peak bagging model. This novel approach is therefore of great interest for possible future automatization of the entire analysis technique.