Predictions of the Nancy Grace Roman Space Telescope Galactic Exoplanet Survey. III. Detectability of Giant Exomoons of Wide Separation Giant Planets

TL;DR

This work assesses the Nancy Grace Roman Space Telescope's ability to detect exomoons orbiting wide-separation giant planets via microlensing. By simulating Roman's Galactic Exoplanet Survey with a realistic Galactic model and a hierarchical moon–planet–star lens, the authors quantify moon detectability using inverse ray shooting and planet-plus-moon versus planet-only fits, finding sensitivity to moons down to about $0.02\,M_⊕$ but a total yield of only ~0.5 moons under fiducial assumptions. The study identifies two main detection channels, shows a linear improvement in detectability with higher sampling rates, discusses the impact of multiple moons, and underscores the need for robust triple-lens modeling to exploit Roman's full exomoon potential. It also provides guidance on how survey cadence and field strategy could enhance moon yields in future planning.

Abstract

The Nancy Grace Roman Space Telescope (Roman) will conduct a Galactic Exoplanet Survey (RGES) to discover bound and free-floating exoplanets using gravitational microlensing. Roman should be sensitive to lenses with mass down to ~ 0.02 $M_{\oplus}$, or roughly the mass of Ganymede. Thus the detection of moons with masses similar to the giant moons in our Solar System is possible with Roman. Measuring the demographics of exomoons will provide constraints on both moon and planet formation. We conduct simulations of Roman microlensing events to determine the effects of exomoons on microlensing light curves, and whether these effects are detectable with Roman. We focus on giant planets from 30 $M_{\oplus}$ to 10 $M_{Jup}$ on orbits from 0.3 to 30 AU, and assume that each planet is orbited by a moon with moon-planet mass ratio from $10^{-4}$ to $10^{-2}$ and separations from 0.1 to 0.5 planet Hill radii. We find that Roman is sensitive to exomoons, although the number of expected detections is only of order one over the duration of the survey, unless exomoons are more common or massive than we assumed. We argue that changes in the survey strategy, in particular focusing on a few fields with higher cadence, may allow for the detection of more exomoons with Roman. Regardless, the ability to detect exomoons reinforces the need to develop robust methods for modeling triple lens microlensing events to fully utilize the capabilities of Roman.

Figures (13)

• Figure 1: Geometry of lens systems and their caustics. Each panel shows a different regime of caustic structure, defined using the criteria from Equation \ref{['eq:caustic_topology']}. The star, plus, and square indicate the positions of the star, planet, and moon, respectively. $s_p$ and $s_m$ indicate the distances between the star and planet, and the moon and planet, respectively. $\psi$ indicates the orientation of the moon relative to the planet-star axis. The red lines show the location and shape of the planetary caustic(s), or, in the case of the resonant regime, the resonant caustics. For the close and wide regimes, $x_{cen}$ shows the distance between the star and the center of the caustic. $\Delta\xi$ and $\Delta\eta$ are the vertical and horizontal widths of the planetary caustic, respectively. The cross section of the planetary caustic, $s_{caus}$, is the geometric mean of $\Delta\eta$ and $\Delta\xi$. Note: this figure is not drawn to scale.
• Figure 2: An example magnification map generated using inverse ray-shooting and the following parameters: $q_p=0.0026$, $q_m=0.01$, $s_p=2.058$, $s_m=0.9648$, $\Psi=43^\circ$. On this scale, darker color indicates higher magnification. The moon introduces additional structure to the caustic that is not present with just a planet.
• Figure 3: A histogram of the root-mean-square (rms) deviation between light curves computed using our inverse ray-shooting algorithm and light curves computed using MulensModelPoleski:2019. We use a subset of 2590 light curves computed as part of the simulations described in this work. For the rms calculation, we only use the region of the light curve calculated using ray-shooting and exclude the parts calculated with MulensModel. The gray-filled histogram shows the rms between ray-shooting light curves with only a planet and a MulensModel light curve calculated with the same parameters. The blue hatched histogram uses light curves that include a moon compared to the same MulensModel light curve. The open histogram shows the rms between planet-only ray shooting light curves (no moon), including expected photometric noise (see Section \ref{['sec:romansim']}), compared to the same MulensModel light curve. The blue histogram with the moon is essentially the gray (planet-only) histogram shifted to larger deviations, resulting in a slight excess compared to the planet-only histogram at high rms that comes from the perturbations due to the moons.
• Figure 4: Panel a shows the range of semimajor axes $a_p$ and planet masses $M_p$ for which we simulated companion moons. We assume a $0.3\ M_{\odot}$ host star, which is approximately the mean mass of the stars in our simulation. We assume a distance to the lens, $D_l$ of 7.37 kpc, which is approximately the front of the Galactic Bulge. The black solid lines show contours of constant $s_{\rm Hill} \equiv a_{Hill} / R_{{\rm E},p}$, e.g., the ratio of the Hill radius of the planet to the Einstein ring radius of a planet. The dashed blue curves show contours of constant $s_{caus}$, the mean caustic cross section (See Equations \ref{['eq:s_caustic_close']} and \ref{['eq:s_caustic_wide']}). Note that the definition of $s_{caus}$ for the close and wide regimes are slightly different. For the wide regime, $s_{caus}$ is simply related to the size of the planetary caustic. For the close regime, $s_{caus}$ also takes into account the vertical separation between the two planetary caustics. The gray, shaded region shows the region of the resonant caustic topology that we do not include in our simulations. Each offset panel (b-g) shows the caustic structures (in red) the Hill radius (in black) and the planetary Einstein ring radius (in blue) centered at the location of the planet (the plus symbol). The caustics were calculated using the triplelens package Kuang_2021. The gray, shaded region in the inset is the region around the planet where we simulated moons (0.1-0.5 Hill radii). A moon with a randomly chosen semimajor axis and orbital phase is shown as a square. The arrows in each panel point to the locations of the small caustics from the moon. These panels are connected to the corresponding planet mass and semimajor axis in the main panel. This figure is discussed in more detail at the end of Section \ref{['sec:general']}.
• Figure 5: The masses of large Solar System moons in Earth masses versus their semimajor axes in units of the Hill radius. Each different symbol and color represents the moons of a different planet. The red symbols are the two exomoon candidates, Kepler-1708bi Kipping_2022 and Kepler-1625bi Teachey_2018. The gray points are our simulated moons. Each of the black stars is a detected moon in our simulation.