A new photometric ephemeris for the 2M1510 AB double brown dwarf eclipsing binary system

TL;DR

This work refines the photometric ephemeris of the double brown dwarf eclipsing binary 2M1510 AB by combining seven years of TESS FFI data with ground-based SPECULOOS observations and a robust BLS+MCMC analysis. The authors obtain a highly precise orbital period of $P = 20.897782 \pm 0.000036$ d, along with a well-constrained secondary-eclipse epoch $T_{\mathrm{sec}} = 2457961.53520 \pm 0.00026$ and inclination $i = 88.47\deg \pm 0.03\deg$, improving upon prior measurements and enabling future follow-up. A key result is the tension with a recent Doppler-based solution: after correcting timing offsets in the radial velocities, there is no evidence for apsidal precession or a circumbinary planet, casting doubt on the ABb planet claim. The analysis also discusses a deeper TESS eclipse relative to SPECULOOS and concludes that residual depth differences are not explained by simple filter effects or orbital evolution. Overall, the refined ephemeris facilitates scheduling of future observations (e.g., with JWST) to further probe the system’s substellar physics and potential dynamical effects.

Abstract

Eclipsing brown dwarfs are important calibrators of sub-stellar evolution models used to infer the characteristics of directly imaged brown dwarfs and giant exoplanets. Only two double brown dwarf eclipsing binary systems are known, among them 2MASS J15104786-2818174 (2M1510 AB), published in 2020 with a poorly constrained orbital period. Here we analyse TESS full-frame image (FFI) photometry of this faint ($T=15.9$) binary and detect a significant (${>}10σ$) periodic signal spanning TESS Cycles 1-7, consistent with previous data. We refine the orbital period to $20.897782 \pm 0.000036$ d, reducing its present-day uncertainty from 18 h to 8 min. Our work is crucial for scheduling follow-up observations of this system for detailed study with other photometric facilities. We also find that a recent orbital solution from Doppler data is inconsistent with existing photometry. A timing offset in the Doppler data may have produced a spurious signal mimicking retrograde apsidal precession, from which the claimed circumbinary planet 2M1510 ABb was inferred. From our best attempt at correcting the data we were unable to reconcile the radial velocity data with the photometry, suggesting that the radial velocity uncertainties are underestimated, and that the circumbinary planet 2M1510 ABb may be a false positive.

Figures (3)

• Figure 1: Lightcuves of 2M1510AB observed by TESS. Solid red line indicates position of transits predicted from triaud2020. Further red shaded region indicates its $1\sigma$ uncertainty.
• Figure 2: Periodograms from Box Least Squares search.a) is the periodogram from the BLS search using TESS and SPECULOOS. b is magnified for 5 $\sigma$ around the period predicted by triaud2020. c and d, same as above, using only TESS data. Red arrows and red line indicate the period predicted by triaud2020. Further red shaded region around the red line indicates 1 $\sigma$ uncertainty.
• Figure 3: Fits for TESS and SPECULOOS data.a), b), c) and d) show the four identified eclipses in the TESS data, binned to 15 minutes. e) shows the eclipse in the SPECULOOS data, and f) shows the phase-folded TESS data of the 4 eclipses, binned to 15 minutes and centred on the mid-eclipse. The red line in each figure indicates the best fit solution from the MCMC run, with a further red shaded region indicating its 1 $\sigma$ uncertainty. The black dotted line in f) represents the best fit solution whilst fixing the surface brightness ratio at 0.827 (i.e. same as SPECULOOS).