A Deep Search for Exomoons Around WISE 0855 With JWST

TL;DR

This work probes the existence of exomoons around the free-floating planet WISE 0855 using 11 hours of JWST/NIRSpec time-series spectroscopy. The authors implement a two-wavelength, variability-aware transit search combining Gaussian Process modeling with an identical, concurrent transit signature in both light curves, and they calibrate their detection via injection-recovery tests. No statistically significant exomoon is found, with a 0.43 in Δlogz_final for the GP+transit vs GP-only models, and injection tests show high recoverability for transits of depth ≥0.5% (≈ Titan-analog), while shallower transits become increasingly challenging under host variability. They conclude JWST can constrain exomoon occurrence around FFPs, estimating ~91–95% chances of detecting Titan- or Io-like moons after ~18 similar observations, thereby paving the way for population-level constraints on exomoon frequencies in the substellar regime.

Abstract

JWST is collecting time-series observations of many free-floating planets (FFPs) to study their weather, but these light curves are the ideal datasets to search for exomoons that transit the FFP during observations. In this paper, we present observations of the planetary-mass Y dwarf ($T=250-285K$, $M = 6.5\pm3.5 M_{Jup}$, d = 2.3$\,$pc) WISE J085510.83-071442.5 (WISE 0855), whose proximity and brightness make it ideal for a transiting exomoon search. We examine 11 hours of time-series spectra from the JWST Near-Infrared Spectrograph (NIRSpec) whose sensitivity, in combination with Gaussian process (GP) modeling, allows for the disentanglement of exomoon transits from WISE 0855's variability. We do not find statistically significant evidence of an exomoon transit in this dataset. Using injection and recovery tests of artificial transits for depths ranging between 0.1-1% (0.35-1.12 $R_{\oplus}$) we explore the exomoon parameter space where we could successfully detect transits. For transit depths $\geq 0.5\%$ (1.96$\,R_{\text{Titan}}$), our detection rate is 96%, which, for WISE 0855, corresponds to a moon with a companion-to-host mass ratio similar to that of Titan and Saturn. Given our sensitivity, transit probabilities, and our observational duration, we determine a $\sim$91% probability of detecting a Titan mass analog exomoon after 18 such observations if every observed system hosts a Titan mass analog exomoon in a Galilean-like system. This suggests that JWST observations of dozens of FFPs could yield meaningful constraints on the occurrence rate of exomoons. This paper is the first demonstration that JWST is sensitive to Galilean moon mass analogs around FFPs.

Figures (18)

• Figure 1: Combined Spectrum of the 11-hour observation of WISE 0855 created from averaging the 44 spectra. The error bars are too small to be visible.
• Figure 2: Probability of an exomoon transit during the 11 hour WISE 0855 observation assuming a random orbital inclination angle. Different combinations of predicted radius and mass values for WISE 0855 are shown. The black line for each moon represents the extent of possible transit probabilities given these various mass and radius uncertainties for the host. The largest Roche limit for an Io-like moon around the densest prediction of WISE 0855 is at 0.00177 AU. Shown in gray is the distance range within which moons would be tidally disrupted.
• Figure 3: Each curve represents transit duration for an edge-on system as a function of separation for a range of possible mass and radius values for WISE 0855 with the separations of the Galilean moons shown for reference. The black line emphasizes the extent of possible transit durations for each moon given these various radius and mass combinations.
• Figure 4: The time-series spectra with the full wavelength range (in microns) of the observations shown on the x-axis, observation time (in hours) on the y-axis, and a color bar representing the amplitude variability for any given flux value as a percent deviation from the mean. The 2 wavelength regions used to construct the light curves are shown where each has a distinct variability pattern. Region 1: negative, positive, negative positive amplitude; Region 2: negative, positive amplitude. The picket fence pattern (narrow striping) visible in region 2 is due to CO absorption.
• Figure 5: The light curves, normalized to 1, of regions 1 and 2 built from a weighted average calculated using empirical errors.