High Contrast Imaging with Spitzer : Constraining the Frequency of Giant Planets out to 1000 AU separations

TL;DR

This study re-analyzed archival Spitzer data to place new constraints on the frequency of giant planets at wide separations (100-1000 AU), finding no detections and limiting their occurrence to less than 9%.

Contribution

It applies advanced high-contrast imaging techniques to Spitzer data to systematically constrain the population of giant planets at very wide separations, a previously underexplored parameter space.

Findings

No planetary detections in the sample.

Constrained the occurrence rate of 0.5-13 M_J planets at 100-1000 AU to less than 9%.

Demonstrated the effectiveness of high-contrast techniques on archival infrared data.

Abstract

We report results of a re-analysis of archival Spitzer IRAC direct imaging surveys encompassing a variety of nearby stars. Our sample is generated from the combined observations of 73 young stars (median age, distance, spectral type = 85 Myr, 23.3 pc, G5) and 48 known exoplanet host stars with unconstrained ages (median distance, spectral type = 22.6 pc, G5). While the small size of Spitzer provides a lower resolution than 8m-class AO-assisted ground based telescopes, which have been used for constraining the frequency of 0.5 - 13 $M_J$ planets at separations of $10 - 10^2$ AU, its exquisite infrared sensitivity provides the ability to place unmatched constraints on the planetary populations at wider separations. Here we apply sophisticated high-contrast techniques to our sample in order to remove the stellar PSF and open up sensitivity to planetary mass companions down to 5\arcsec\ separations. This enables sensitivity to 0.5 - 13 $M_J$ planets at physical separations on the order of $10^2 - 10^3$ AU , allowing us to probe a parameter space which has not previously been systematically explored to any similar degree of sensitivity. Based on a colour and proper motion analysis we do not record any planetary detections. Exploiting this enhanced survey sensitivity, employing Monte Carlo simulations with a Bayesian approach, and assuming a mass distribution of $dn/dm \propto m^{-1.31}$, we constrain (at 95% confidence) a population of 0.5 - 13 $M_J$ planets at separations of 100 - 1000 AU with an upper frequency limit of 9%.