Those of you who have been following the controversy over the dimming of KIC 8462852 (Tabby’s Star) may remember an interesting note at the end of Bradley Schaefer’s last post on Centauri Dreams. Schaefer (Louisiana State University) had gone through his reasoning for finding a long-term dimming of the star in the DASCH (Digital Access to a Sky Century@Harvard) database. His third point about the star had to do with the work of Ben Montet (Caltech) and Joshua Simon (Carnegie Observatories).

Montet and Simon’s work relied on an interesting premise. Tabby’s Star had been discovered because it was in the Kepler field, and thus we had high-quality data on its behavior, the unusual light curves that the Planet Hunters team brought to the attention of Tabetha Boyajian. As the researchers note in a new paper, Kepler found ten significant dips in the light curve over the timespan of the Kepler mission, dips that were not only aperiodic but irregular in shape, and that varied enormously, from fractions of one percent up to 20% of the total flux of KIC 8462852.


Image: Montage of flux time series for KIC 8462852 showing different portions of the 4-year Kepler observations with different vertical scalings. Panel ‘(c)’ is a blowup of the dip near day 793, (D800). The remaining three panels, ‘(d)’, ‘(e)’, and ‘(f)’, explore the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Credit: Boyajian et al., 2015.

Schaefer noted in his Centauri Dreams post (see Further Thoughts on the Dimming of KIC 8462852) that if Tabby’s Star were actually fading at a rate of 0.164 mag/cen, then it should have undergone fading during the period it was under observation by Kepler (in fact, it should have faded by 0.0073 mag over the Kepler lifetime on the main Cygnus field). Montet and Simon have now presented us with their analysis in a paper just up on the arXiv server.

A fading of the kind Schaefer described would be well above the photometric precision of the Kepler instrument. Montet and Simon realized they could search for long-term trends by using the full-frame images (FFI) collected during the Kepler mission. Eight of these were recorded at the beginning of the mission, with another FFI recorded each month throughout the mission. Given that the mission lasted four years, a star dimming at the rate Schaefer suggests should decrease in brightness by 0.6% over the Kepler baseline. And as the authors point out, using FFI data avoids the removal of the dimming trend by the data processing pipeline.

The results: The study, which worked with KIC 8462852 and seven nearby comparison stars, found that in the first three years of the Kepler mission, Tabby’s Star dimmed at a rate of 0.341%±0.041% per year. Over the next six months, it decreased in brightness by 2.5%, and then stayed at that level during the duration of the primary Kepler mission. The paper continues:

We then compare this result to a similar analysis of other stars of similar brightness on the same detector, as well as stars with similar stellar properties, as listed in the KIC, in the Kepler field. We find that 0.5% of stars on the same detector and 0.7% of stars with similar stellar properties exhibit a long-term trend consistent with that observed for KIC 8462852 during the first three years of the Kepler mission. However, in no cases do we observe a flux decrement as extreme as the 2.5% dip observed in Quarters 12-14 of the mission. The total brightness change of KIC 8462852 is also larger than that of any other star we have identified in the Kepler images.

Screenshot from 2016-08-05 11:11:46

Image: Photometry of KIC 8462852 as measured from the FFI data. The four colors and shapes (green squares, black circles, red diamonds, and blue triangles) represent measurements from the four separate channels the starlight reaches as the telescope rolls. The four subpanels show flux from each particular detector individually. The main figure combines all observations together; we apply three linear offsets to the data from different channels to minimize the scatter to a linear fit to the first 1100 days of data. In all four channels, the photometry is consistent with a linear decrease in flux for the first three years of the mission, followed by a rapid decrease in flux of ? 2.5% over the next six months. The light gray curve represents one possible Kepler long cadence light curve consistent with the FFI photometry created by fitting a spline to the FFI photometry as described in Section 4. The large dips observed by Boyajian et al. (2016) are visible but narrow relative to the cadence of FFI observations. The long cadence data behind this figure are available online. Credit: Montet & Simon.

M. A. Thompson (University of Hertfordshire) and colleagues published a recent study in Monthly Notices of the Royal Astronomical Society reporting their findings using millimetre and sub-millimetre photometry. The paper finds that a dust cloud orbiting Tabby’s Star would have to be no larger than 7.7 Earth masses of material within a radius of 200 AU, adding “Such low limits for the inner system make the catastrophic planetary disruption hypothesis unlikely.”

Montet and Simon don’t necessarily agree, but in any case there are other problems. The authors think the light curve is “…consistent with the transit of a cloud of optically thick material orbiting the star,” and that such a cloud could be small enough to meet Thompson and team’s requirements. The breakup of a small body or a recent collision producing a large dust cloud could also produce a cometary family that transited the host star as a single group. But we’re still not out of the woods:

To explain the transit ingress timescale, the cloud would need to be at impossibly large distances from the star or be slowly increasing in surface density. The flat bottom of the transit would then suggest a rapid transition into a region of uniform density in the cloud, which then continues to transit the star for at least the next year of the Kepler mission. Moreover, such a model does not naturally account for the long-term dimming in the light curve observed in both DASCH and the Kepler FFI data, suggesting that this idea is, at best, incomplete.

A deeply mysterious star, our KIC 846285. Montet and Simon call for alternative hypotheses and new data to help us explain existing observations, and we can be glad to have Tabetha Boyajian’s team on the case thanks to the success of the recent Kickstarter campaign. Observations are already in progress at the Las Cumbres Observatory Global Telescope Network, and the Kickstarter funds will take us deep into 2017. For more on the Las Cumbres work, see Corey Powell’s recent interview with Boyajian for Discover Magazine, from which this:

From our new observations, we’ll be able to tell a lot about the material that’s passing in front of the star: if it’s some kind of dusty thing, some kind of solid thing. [Boyajian’s working hypothesis is that the dimming is caused by a huge swarm of comets, set loose perhaps by some cataclysmic event around the star.] What’s also important is that we will also get a baseline of spectral observations so we can look at if there’s any radial velocity shift or if there’s any variable emission of the lines, things we’d expect comets to have.

The paper is Montet and Simon, “KIC 8462852 Faded Throughout the Kepler Mission,” submitted to the AAS Journals and available as a preprint. The Thompson paper on circumstellar dust in this system is “Constraints on the circumstellar dust around KIC 8462852,” published online by Monthly Notices of the Royal Astronomical Society 25 February 2016.