Photon flux time series from the Kepler space observatory for star KIC 8462852. Graphic: Boyajian, et al., 2015
Photon flux time series from the Kepler space observatory for star KIC 8462852. Graphic: Boyajian, et al., 2015

[UPDATE5: In SETI: No Signal Detected from KIC 8462852, Paul Gilster quotes Seth Shostak: “The history of astronomy tells us that every time we thought we had found a phenomenon due to the activities of extraterrestrials, we were wrong. But although it’s quite likely that this star’s strange behavior is due to nature, not aliens, it’s only prudent to check such things out.” Well said, and I do hope SETI succeeds in my lifetime. –Jim]

By Jim Galasyn
23 October 2015

(Desdemona Despair) – The last couple of weeks have seen a flurry of speculation about an observation made by the Kepler space observatory: is it evidence of alien megastructures, like a “Dyson swarm”? The Kepler satellite looks for extrasolar planets (“exoplanets”) circling distant stars in a small square of sky near the northern constellations of Cygnus, Draco, and Lyra. Launched in 2009, the Kepler mission has been immensely successful, discovering more than 1,000 exoplanets in about 440 stellar systems by the year 2015. Kepler is an orbiting telescope that looks for the slight dimming of a star’s light as a planet passes in front of the stellar disk. When a planet transits its parent star, the light from the star dims, usually in a symmetrical pattern.

Transit method of detecting extrasolar planets. Graphic: Wikipedia
Transit method of detecting extrasolar planets. Graphic: Wikipedia
Transit light curves for Kepler 4b, Kepler 5b, Kepler 6b, Kepler 7b, and Kepler 8b. Illustration from Bill Borucki's Jan 2010 AAS Presentation. Graphic: Bill Borucki
Transit light curves for Kepler 4b, Kepler 5b, Kepler 6b, Kepler 7b, and Kepler 8b. Illustration from Bill Borucki’s Jan 2010 AAS Presentation. Graphic: Bill Borucki

How then to explain these asymmetrical light curves from the star KIC 8462852?

Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Graphic: Boyajian, et al., 2015 / Mon. Not. R. Astron. Soc.
Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Graphic: Boyajian, et al., 2015 / Mon. Not. R. Astron. Soc.

Tabetha Boyajian and colleagues have examined the unusual photon flux data from KIC 8462852 (“Tabby’s star”) in various ways, and they conclude:

The scenario most consistent with the data in hand is the passage of a family of exocomet fragments, all of which are associated with a single previous breakup event. [“Planet Hunters X. KIC 8462852 – Where’s the Flux?”]

It’s an unsatisfying explanation for a number of reasons that they discuss in their paper. Naturally, people have started wondering if we’re looking at the long-sought evidence of extraterrestrial intelligence: the shadows of giant, irregular structures created by an ancient, spacefaring civilization. I’m as fascinated with SETI as the next geek, but claims of evidence for extraterrestrials must be viewed with great skepticism. The photon flux data captured by Kepler appear to be unlike anything seen before, but really, it would be incredibly lucky for one of the very first probes sent by humans on a planet-finding mission to discover evidence of an alien civilization. So a closer look for natural causes is necessary.

After obsessing over the graphs in the Boyajian paper for a while, I ran across a paper posted by commenter Michael at Centauri Dreams: “Measurement of spin-orbit misalignment and nodal precession for the planet around pre-main-sequence star PTFO 8-8695 from gravity darkening” (Barnes, et al., 2013) [pdf]. This paper goes a long way toward explaining the anomalous light curves of KIC 8462852. [But see UPDATE2 below.]

The idea is that some stars don’t have uniformly bright disks. Some stars spin at a high rate, giving them an oblate spheroidal shape and causing an effect called “gravity darkening”.

When a star is oblate, it has a larger radius at its equator than it does at its poles. As a result, the poles have a higher surface gravity, and thus higher temperature and brightness. Thus, the poles are “gravity brightened”, and the equator “gravity darkened”.[1] The star becomes oblate (and hence gravity darkening occurs) because the centrifugal force resulting from rotation creates additional outward pressure on the star. [Gravity darkening]

CHARA/MIRC imaging of 6 rapidly rotating stars, spanning a wide spectral range from B8 to F2. The stars are scaled relatively to their linear sizes. The stars are oblate due to their rapid rotation. The polar areas of these stars are bright and their equatorial areas are dark because of the gravity darkening effect. Graphic: Ming Zhao
CHARA/MIRC imaging of 6 rapidly rotating stars, spanning a wide spectral range from B8 to F2. The stars are scaled relatively to their linear sizes. The stars are oblate due to their rapid rotation. The polar areas of these stars are bright and their equatorial areas are dark because of the gravity darkening effect. Graphic: Ming Zhao

There are four discrete events in the Kepler data for KIC 8462852, and planetary transits across a gravity-darkened disk are plausible causes for all of them. The first event occurs at Kepler day 792 into the mission (D800). It’s similar to the usual light curve for a transiting planet, but it’s asymmetrical. Boyajian says of the first event:

Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dip which occurs near day 793 (D800). Graphic: Boyajian, et al., 2015 / Mon. Not. R. Astron. Soc.
Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dip which occurs near day 793 (D800). Graphic: Boyajian, et al., 2015 / Mon. Not. R. Astron. Soc.

A fairly generic prediction of transits of comet-like bodies may be that their light-curves show signs of their tails. The light-curve expected for a typical event then has a relatively fast ingress as the head of the comet passes in front of the star, but a slower egress as the tail passes (e.g. Lecavelier Des Etangs, et al., 1999; Rappaport, et al., 2012). However, the D800 event [what I’m calling day 792 –Jim] shows the opposite (see panel ‘c’ in Figure 1). Possible resolutions of this issue are that the D800 comet fragment received a large kick with an orientation that sheared it out in such a way to form a “forward tail”. Such forward comet tails produced by the fragments being kicked toward the star have been studied in the literature, but require the tail to be large enough to overcome the effects of radiation pressure (Sanchis-Ojeda, et al., 2015). Alternatively, this event could be comprised of two dips superimposed to have the appearance of a forward tail. While several issues remain to be explored, of the scenarios considered we conclude that a cometary origin seems most consistent with the data to hand.

But a similar asymmetrical light curve can be produced by a transit across a gravity-darkened disk:

Transit lightcurve shape and graphic depiction of what the transit event at star PTFO 8-8695 might have looked like at the 2011 observation. Graphic: Barnes, et al., 2013 / The Astrophysical Journal
Transit lightcurve shape and graphic depiction of what the transit event at star PTFO 8-8695 might have looked like at the 2011 observation. Graphic: Barnes, et al., 2013 / The Astrophysical Journal

The third event at day 1540 is symmetrical but features lobes on either side of the minimum.

Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Graphic: Boyajian, et al., 2015 / Mon. Not. R. Astron. Soc.
Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Graphic: Boyajian, et al., 2015 / Mon. Not. R. Astron. Soc.

My first thought was that this is clearly a ring system, with a “Cassini division” on either side of the central planet. But a similar curve results from a transit across a symmetrically brightened disk:

Transit lightcurve shape and graphic depiction of what the transit event at star PTFO 8-8695 might have looked like at the 2010.4 interpolated values. Graphic: Barnes, et al., 2013 / The Astrophysical Journal
Transit lightcurve shape and graphic depiction of what the transit event at star PTFO 8-8695 might have looked like at the 2010.4 interpolated values. Graphic: Barnes, et al., 2013 / The Astrophysical Journal

The second and fourth events at day 1520 and day 1570 have a qualitatively similar shape but different magnitude. I’ve applied an arbitrary horizontal (time-axis) scaling to size the day 1570 event roughly with the day 1520 event.

Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur near day 1520 and day 1570. The curve for day 1570 has been scaled along the horizontal (time-axis) to show similarities with the day 1520 curve. Adapted from Boyajian, et al., 2015. Graphic: James P. Galasyn
Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur near day 1520 and day 1570. The curve for day 1570 has been scaled along the horizontal (time-axis) to show similarities with the day 1520 curve. Adapted from Boyajian, et al., 2015. Graphic: James P. Galasyn

Although they have interesting high-frequency details, in general both curves follow the characteristic shape of a transit across a brightened pole.

Transit lightcurve shape and graphic depiction of what the transit event at star PTFO 8-8695 might have looked like at the 2010.1 interpolated values. Graphic: Barnes, et al., 2013 / The Astrophysical Journal
Transit lightcurve shape and graphic depiction of what the transit event at star PTFO 8-8695 might have looked like at the 2010.1 interpolated values. Graphic: Barnes, et al., 2013 / The Astrophysical Journal

The two events appear to be transits of two planets. The first planet is large, occluding 20% of the star’s disk. The second planet is much smaller, occluding just 8% of the stellar disk. Because the second event has shorter duration, the smaller planet is moving faster and therefore is orbiting closer to the star. The high-frequency variations aren’t predicted in the Barnes paper. I speculate that the gravity darkening around the star’s equator may be banded, like Jupiter’s atmosphere. If this idea holds up, planetary transits could be a valuable tool for inferring stellar surface dynamics. [But see UPDATE below.] Together, the three transits might look something like this:

Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Annotations show hypothetical planetary transits that shape the light curves. Adapted from Boyajian, et al., 2015 and Barnes, et al., 2013. Graphic: James P. Galasyn
Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dips which occur during the 90-day interval from day 1490 to day 1580 (D1500). Annotations show hypothetical planetary transits that shape the light curves. Adapted from Boyajian, et al., 2015 and Barnes, et al., 2013. Graphic: James P. Galasyn

As noted in Wright, et al. (2015), there are other interesting events in the KIC 8462852 data:

Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dip which occurs near day 262. Graphic: Wright, et al., 2015
Photon flux time series from the Kepler space observatory for star KIC 8462852 showing the dip which occurs near day 262. Graphic: Wright, et al., 2015

The event on the left, at day 262, is an ordinary transit. On the right, at day 1206, is another transit across a symmetrically brightened disk. Both are very small planets, relative to the planet in the day 1520 event. Did Kepler discover evidence of an extraterrestrial civilization? Before we invoke aliens, we should consider the plausible natural explanation provided by Barnes: “An oblique transit path across a gravity-darkened, oblate star leads to the long transit duration and asymmetric lightcurve evident in the photometric data.” UPDATE: A quick literature search produced another interesting paper on stellar system PTFO 8-8695, “Revisiting a gravity-darkened and precessing planetary system PTFO 8-8695: spin-orbit non-synchronous case” (Kamiaka, et al., 2015)[pdf]. Here are three possible transits that could fit the observed data:

The evolution of PTFO 8-8695 transit lightcurves for three possible solutions from the 2009 to 2010 observational epochs. Graphic: Kamiaka, et al., 2015 / Astronomical Society of Japan
The evolution of PTFO 8-8695 transit lightcurves for three possible solutions from the 2009 to 2010 observational epochs. Graphic: Kamiaka, et al., 2015 / Astronomical Society of Japan

UPDATE3: Here are the interpolated curves from the Kamiaka paper, compared with the observed data from KIC 8462852.

Light curves for stars PTFO8-8695 and KIC8462852. The middle 3 curves for PTFO 8-8695 are interpolations of orbits that fit the data in the left and right panels. From Kamiaka, et al, 2015 and Boyajian, et al., 2015. Graphic: James P. Galasyn
Light curves for stars PTFO8-8695 and KIC8462852. The middle 3 curves for PTFO 8-8695 are interpolations of orbits that fit the data in the left and right panels. From Kamiaka, et al, 2015 and Boyajian, et al., 2015. Graphic: James P. Galasyn

I had thought that bands in the stellar disk would be necessary to account for the high-frequency details of the light curve, but it seems that they’re not.

UPDATE2: One of the anonymice points to a compelling new paper, “Tests of the planetary hypothesis for PTFO 8-8695b” (Yu, et al., 2015)[pdf] that disconfirms the Barnes paper: “All these observations cast doubt on the planetary hypothesis, and suggest instead that the fading events represent starspots, eclipses by circumstellar dust, or occultations of an accretion hotspot.”

The Yu paper finds that the hypothetical curve fits computed by Barnes, et al., don’t fit the long-term observations of the PTFO 8-8695 system. This is an entirely reasonable result. I don’t think this weakens the case for KIC 8462852, since these wacky curves are present in the observed data.

The other objection to the gravity-darkening hypothesis is that the two big KIC 8462852 events (20% dimming and 8% dimming) are far too large to be caused by planets; a comparable Jupiter transit would dim the sun by about 1%. This is a fair point, but it doesn’t significantly weaken the case for transits of some large object(s) across a gravity-darkened disk. Also, it’s worth noting that one of the solutions presented by Kamiaka (the blue trace in the above graphs) predicts a dimming of 30% for a planet in the 3-4 Jupiter mass range.

[UPDATE4] I emailed Professor Barnes for his thoughts on KIC 8462852, and here’s his reply:

Indeed, the individual transit events in the KIC 8462852 lightcurve definitely show the signature of a misaligned object transiting a gravity-darkened stellar disk.  Even if PTFO 8-8695b looks like it probably isn’t a planet, we have seen this kind of gravity-darkening signature at Kepler-13b, for instance.  But WHAT might be transiting at such odd cadences and with such differing durations is far less clear. The Kepler-13 gravity-darkened transit signature paper can be found here.

Thank you, Professor Barnes!