Predicting missing planets in multiplanet system populations via analytical assessments of dynamical packing

TL;DR

This paper introduces an analytical method to identify potential missing planets in multi-planet systems by assessing their dynamical packing, applying it to Kepler data, and estimating the number and characteristics of undetected planets.

Contribution

The paper presents a novel analytical approach to predict potential undiscovered planets in multi-planet systems based on dynamical spacing criteria, validated with Kepler data.

Findings

Up to 560 planet pairs could host additional planets.

164 pairs have a high probability of being unpacked.

Approximately 28.2% of these potential planets could be Earth-sized or smaller.

Abstract

We present a new analytical method to identify potential missed planets in multiplanet systems found via transit surveys such as those conducted by Kepler and TESS. Our method depends on quantifying a system's dynamical packing in terms of the dynamical spacing $\Delta$, the number of mutual Hill radii between adjacent planets ("planet pair"). The method determines if a planet pair within a multi-planet system is dynamically unpacked and thus capable of hosting an additional intermediate planet. If a planet pair is found to be unpacked, our method constrains the potential planet's mass and location. We apply our method to the Kepler primary mission's population of 691 multi-candidate systems, first via direct calculations and then via Monte Carlo (MC) analysis. The analysis was repeated with three proposed values from the literature for minimum $\Delta$ required for planet pair orbital stability ($\Delta = 10$, $12.3$, and $21.7$). Direct calculations show that as many as $560$ planet pairs in $691$ Kepler multi-candidate systems could contain additional planets ($\Delta = 12.3$). The MC analysis shows that $164$ of these pairs have a probability $\geq 0.90$ of being unpacked. Furthermore, according to calculated median mass efficiencies calculated from packed Kepler systems, $28.2\%$ of these potential planets could be Earths and Sub-Earths. If these planets exist, the masses and semimajor axes predicted here could facilitate detection by characterizing expected detection signals. Ultimately, understanding the dynamical packing of multi-planet systems could help contribute to our understanding of their architectures and formation.