This is somewhat a sad story for those who believe in rapid advancement of astronomy science. Today a team of scientists submitted a paper to arXiv.org, in which they elaborate on the topic of the seven-planet system Kepler-90 (KOI-351) and the possibility of existence of exomoon orbiting around the host star of this system.
The authors of this scientific analysis note, that despite nearly 2,000 exoplanets having now been discovered, nobody has detected any satellite orbiting these bodies, at least yet. And the motivation for the discovery of the first exomoon is as great as ever: some of the detected exoplanets are even smaller than Mercury, so there is a viable ground to believe that a possibility to discover satellites is very real.
This belief is also strongly supported by theoretical science which states that state-of-the-art gravitational microlensing and transit photometry is sensitive enough to ‘large’ exomoons. By term ‘large’ the scientists mean bodies with masses approaching one tenth of the Earth mass. That’s not enough to discover something like our Moon (the Moon-Earth mass ratio is 0.0123), but could be enough to detect larger satellites.
The seven-planet transiting system Kepler-90 was discovered in 2013. The same year, one of the teams behind the discovery raised a theory that the planet Kepler-90g could host a large exomoon. This claim was partially supported by the presence of a moon-like transit signal appearing in the data of the third transit of this gas giant, Kepler-90g.
At that time scientists behind this discovery remained cautious and did not claim to have discovered the first exomoon candidate. Meanwhile, the moon-like signal remained a mystery which prompted other researchers to look for plausible explanations of this phenomenon.
The authors of the current study analyzed a full set of photometric observations of Kepler-90g with aim to determine patterns resembling planet-moon configuration. The main clues negating the exomoon hypothesis were: 1) moon-like signal was detected only during a single transit of Kepler-90g out of six, so the satellite had to ‘hide” somewhere during other observations, and 2) the transit of the assumed satellite appeared ~21.5 hours later than that of Kepler-90g, implying a large planet-moon separation.
At first, the team performed the light curve analysis. The results indicated that the observed light curve could be well-explained by Kepler-90g hosting a single large exomoon. The second intermediate conclusion was that the assumed satellite could be within the Hill sphere of the Kepler-90g, i.e. inside of the zone where the gravitational attraction of the planet can retain the ‘control’ of the satellite. So a deeper analysis was required.
The centroid analysis was then performed on the available Kepler data. The transit photometric data was weighed against the signal-to-noise ratio of the transit signal using a technique dubbed by the authors as ‘transit centroid analysis’. The centroid of the implied exomoon was highly delocalized, so the team decided to substitute available data with a fake transit signal. The fake signal was calculated using light curves of Kepler-90g’s transit by reducing the magnitude of the points to match the depth of the potential satellite and then injecting this data in place of the original moon-like signal.
This test confirmed that the moon-like transit was not localized at the expected position, and this result was not due to the low amplitude of the signal, the authors state. They also suggest that the moon-like transit was likely an instrumental artifact, possibly caused by interfering cosmic rays.
Written by Alius Noreika