M-band Imaging of the HR 8799 Planetary System Using an Innovative LOCI-based Background Subtraction Technique

TL;DR

This paper reports deep M-band imaging of the HR 8799 planetary system using an innovative LOCI-based background subtraction technique, enabling detection of three planets and providing new insights into their atmospheres.

Contribution

Development of a novel LOCI-based background subtraction method that significantly improves high-contrast imaging efficiency in M-band observations.

Findings

Detected three HR 8799 planets at M-band for the first time.

Extended photometric coverage to about 5 microns.

Confirmed planets' atmospheric properties are consistent with low gravity and non-equilibrium chemistry.

Abstract

Multi-wavelength observations/spectroscopy of exoplanetary atmospheres are the basis of the emerging exciting field of comparative exoplanetology. The HR 8799 planetary system is an ideal laboratory to study our current knowledge gap between massive field brown dwarfs and the cold 5-Gyr old Solar system planets. The HR 8799 planets have so far been imaged at J- to L-band, with only upper limits available at M-band. We present here deep high-contrast Keck II adaptive optics M-band observations that show the imaging detection of 3 of the 4 currently known HR 8799 planets. Such detections were made possible due to the development of an innovative LOCI-based background subtraction scheme that is 3 times more efficient than a classical median background subtraction for Keck II AO data, representing a gain in telescope time of up to a factor of 9. These M-band detections extend the broad band photometric coverage out to about 5 microns and provide access to the strong CO fundamental absorption band at 4.5microns. The new M-band photometry shows that the HR 8799 planets are located near the L/T-type dwarf transition, similar to what was found by other studies. We also confirm that the best atmospheric fits are consistent with low surface gravity, dusty and non-equilibrium CO/CH4 chemistry models.