When we at WASP search for candidate transit dips we usually ignore any star that we think is a red giant star. Such stars have evolved off the Main Sequence and the outer layers are becoming bloated and puffed-up to many times their original size. Because of this there is no room for a hot-Jupiter planet in a typical orbit of 2 to 4 days, it would be inside the star. Thus if we see a short-period dip apparently on a red giant, it usually means that the system is blended, and that the dip is caused by a fainter, unseen binary.
However, a team led by Jorge Lillo-Box at the Centro de Astrobiología in Madrid have reported that the Kepler “object of interest” KOI-2133 — now promoted to being called Kepler-91 — is a red giant orbited by a Jupiter-mass planet on an orbit of 6.2 days.
This star is the biggest known to host a planet, with a bloated radius of 6.3 times that of the Sun. The planet orbits just outside, and, since the red giant star is still expanding, the authors estimate that it will be engulfed within 50 million years, a very short time compared to usual planet lifetimes. Viewed from the planet the star would fill nearly half the sky.
So puffed up and tenuous are the star’s outer layers, and so close are they to the planet, that the planet’s presence distorts them into an ellipsoidal shape. This produces a modulation of the star’s light at the planet’s orbital period, a modulation that had previously led to Kepler-91 being considered to be a binary star, not a planet host.
The WASP cameras have also been watching Kepler-91 over the years, accumulating 20,000 data points, but could never detect the dip caused by Kepler-91b. Not only is Kepler-91 relatively faint, at V = 12.9, but the transit dip is very shallow.
A typical hot-Jupiter planet transiting a typical star produces a dip of about 1% in the light, which is detectable by WASP. However with Kepler-91 being bloated to 6.3 solar radii its surface area will be 40 times bigger, and hence the planet occults only 1/40th as much of the light, producing a vastly shallower dip. The transit depth in Kepler-91 is only 0.04%, detectable by the superb Kepler photometry from the stability of space, but not detectable from the ground, another reason why WASP ignores stars that are giants.