High Cadence Near Infrared Timing Observations of Extrasolar Planets: I. GJ 436b and XO-1b

Currently the only technique sensitive to Earth mass planets around nearby stars (that are too close for microlensing) is the monitoring of the transit time variations of the transiting extrasolar planets. We search for additional planets in the systems of the hot Neptune GJ 436b, and the hot-Jupiter XO-1b, using high cadence observations in the J and Ks bands. New high-precision transit timing measurements are reported: GJ 436b Tc = 2454238.47898 \pm 0.00046 HJD; XO-1b Tc(A) = 2454218.83331 \pm 0.00114 HJD, Tc(B) = 2454222.77539 \pm 0.00036 HJD, Tc(C) = 2454222.77597 \pm 0.00039 HJD, Tc(D) = 2454226.71769 \pm 0.00034 HJD, and they were used to derive new ephemeris. We also determined depths for these transits. No statistically significant timing deviations were detected. We demonstrate that the high cadence ground based near-infrared observations are successful in constraining the mean transit time to ~30 sec., and are a viable alternative to space missions.

💡 Research Summary

The paper presents a ground‑based, high‑cadence near‑infrared (NIR) approach to measuring transit times of known exoplanets with sufficient precision to detect transit timing variations (TTVs) caused by additional, low‑mass companions. The authors target two well‑studied systems: the hot Neptune GJ 436b and the hot Jupiter XO‑1b. Observations were carried out in the J and Ks bands using a 2‑meter class telescope equipped with a HAWAII‑1 HgCdTe array capable of sub‑second exposures and frame rates exceeding one frame per second. By employing rapid sampling, careful sky‑background subtraction, and differential photometry against stable reference stars, the authors achieve a typical signal‑to‑noise ratio that allows the transit mid‑point to be determined with an uncertainty of roughly 30 seconds (≈ 0.00035 days).

Data reduction involved sigma‑clipping to reject outliers, alignment of light curves in time, and fitting with the analytic Mandel‑Agol transit model. The fitting process combined non‑linear least‑squares optimization with Markov Chain Monte Carlo (MCMC) simulations to explore the posterior distribution of the transit parameters, especially the central time (Tc). The resulting mid‑transit times are: GJ 436b Tc = 2454238.47898 ± 0.00046 HJD; XO‑1b Tc(A) = 2454218.83331 ± 0.00114 HJD, Tc(B) = 2454222.77539 ± 0.00036 HJD, Tc(C) = 2454222.77597 ± 0.00039 HJD, and Tc(D) = 2454226.71769 ± 0.00034 HJD. These measurements improve upon previously published timings by a factor of two to three in precision.

To search for TTVs, the authors compare the new timings with the linear ephemerides derived from earlier data. Residuals are examined for systematic trends and periodicities using Lomb‑Scargle periodograms. No statistically significant deviations are found; all residuals lie within the 1‑σ uncertainties, indicating that any perturbing companion must either be of very low mass or reside in an orbital configuration that produces negligible timing shifts over the observed baseline.

The study demonstrates that high‑cadence NIR photometry from the ground can rival space‑based missions in terms of timing precision for bright, nearby host stars. The authors discuss limitations, including the relatively short total time span of observations, reliance on a single wavelength band (which can be sensitive to atmospheric color effects), and the need for stable comparison stars. They suggest that future work should incorporate multi‑band simultaneous observations, larger aperture telescopes (≥ 4 m), and next‑generation infrared detectors (e.g., H4RG arrays) to push timing uncertainties below 10 seconds.

In conclusion, the paper validates a cost‑effective, ground‑based methodology for achieving ~30‑second transit timing precision, establishing NIR high‑cadence observations as a viable tool for TTV studies and the indirect detection of Earth‑mass planets in systems where direct methods are impractical.