“Peak-to-peak” or “peak-to-trough” amplitudes obtained from Levenberg-Marquardt best fits. In general, for most observations with unconstrained periods, these should be considered minimum amplitudes as the full rotation period was not covered. For the tabulated non-significant detections of variability, the amplitude is taken as the maximum measurable for a given observation.

Given the sometimes rapid light curve evolution of BDs, see e.g. J1324+6358 in Yang et al. (2016), Apai et al. (2017) or J0136+0933 in Artigau et al. (2009), Croll et al. (2016), tabulated periods should be considered to be minimum periods unless well constrained from multiple epochs. Period estimates with uncertainties were obtained from the sinusoidal fit, with the remainder representing where the GLSP crossed the 0.1% false alarm probability (FAP) level.

Median photometric uncertainty for the target (σ_BD) and the average of the median photometric uncertainties for the reference stars used in the final result (σ_RS).

With a FAP of 1.9% and ten targets, we would expect ∼0.2 false positives in the survey and as such regard J2148 to be a highly likely candidate for strong variability. However, for a more robust comparison with the surveys of Radigan et al. (2014) and Vos et al. (2019) we exclude it from the final statistics.