Investigating Flicker noise across the Kepler/K2 sample

[christinahedges]

As discussed in #17 and raised by @kstassun, "Flicker" noise is an excellent way to characterize stellar properties using photometric time series. Flicker noise occurs due to the granulation on the surface of a star, and can be used to measure properties such as the surface gravity. It is possible to identify Flicker noise both in the frequency domain (where it is a "background" to asteroseismic acoustic oscillations) and by fitting in the time domain.

From @ktstassun in #17:
...Bastien et al. (2013, 2016) have shown that from the Kepler long-cadence light curves it is possible to extract the granulation "flicker" that correlates very strongly with stellar surface gravity and thus provides a means for measuring stellar surface gravity with a precision of ~0.1 dex, for stars with logg > 2.5. More recent work demonstrates that the addition of metallicity as a term in the fit enables the surface gravity precision to be improved to ~0.05 dex (Corsaro et al. 2017, Tayar et al. 2018). Application of these methods to the full Kepler + K2 data set holds the promise of enabling the determination of precise stellar properties for stars far beyond the original Kepler footprint. Finally, overlap of the Kepler/K2 sample with upcoming TESS observations should enable calibration and extension of the granulation "flicker" methodology to TESS stars across the entire sky (see, e.g., Stassun et al. 2018).

Applying an analysis of Flicker noise across the entire Kepler sample, perhaps capitalizing on new methods in machine learning, would provide a catalog of robust, independently determined stellar properties using solely the Kepler sample.

[christinahedges]

If anyone has any further comments on this, including how many stars may be amenable to analysis of Flicker noise, and how many of those may or may not have already been analyzed, please do comment here.