Extrasolar planet taxonomy: a new statistical approach

TL;DR

This paper introduces a new statistical classification for extrasolar planets using PCA and clustering, revealing five robust groups that relate to physical parameters and planet formation processes.

Contribution

It develops a novel taxonomy of extrasolar planets based on PCA and hierarchical clustering, linking planetary groups to physical and environmental parameters.

Findings

Identified five robust planetary groups.

Divided hot Jupiters into two main groups.

Found correlations between metallicity, semi-major axis, and eccentricity.

Abstract

In this paper we present the guidelines for an extrasolar planet taxonomy. The discovery of an increasing number of extrasolar planets showing a vast variety of planetary parameters, like Keplerian orbital elements and environmental parameters, like stellar masses, spectral types, metallicity etc., prompts the development of a planetary taxonomy. In this work via principal component analysis followed by hierarchical clustering analysis, we report the definition of five robust groups of planets. We also discuss the physical relevance of such analysis, which may provide a valid basis for disentangling the role of the several physical parameters involved in the processes of planet formation and subsequent evolution. For instance, we were able to divide the hot Jupiters into two main groups on the basis of their stellar masses and metallicities. Moreover, for some groups, we find strong correlations between metallicity, semi-major axis and eccentricity. The implications of these findings are discussed.