No significant correlation between radial velocity planet presence and debris disc properties

Abstract

We investigate whether the tentative correlation between planets and debris discs which has been previously identified can be confirmed at high significance. We compile a sample of 201 stars with known planets and existing far infrared observations. The sample is larger than those studied previously since we include targets from an unpublished Herschel survey of planet hosts. We use spectral energy distribution modelling to characterise Kuiper belt analogue debris discs within the sample, then compare the properties of the discs against a control sample of 294 stars without known planets. Survival analysis suggests that there is a significant ($p\sim 0.002$) difference between the disc fractional luminosity distributions of the two samples. However, this is largely a result of the fact that the control sample contains a higher proportion of close binaries and of later-type stars; both of these factors are known to reduce disc detection rates. Considering only Sun-like stars without close binary companions in each sample greatly reduces the significance of the difference ($p\sim 0.3$). We also find no evidence for a difference in the disc fractional luminosities of stars hosting planets more or less massive than Saturn ($p\sim 0.9$). Finally, we find that the planet hosts have cooler discs than the control stars, but this is likely a detection bias, since the warmest discs in the control sample are also the faintest, and would thus be undetectable around the more distant planet hosts. Considering only discs in each sample that could have been detected around a typical planet host, we find $p\sim 0.07$ for the temperatures.