JWST/MIRI coronagraphic performances as measured on-sky

TL;DR

This paper presents the first on-sky contrast measurements of JWST's MIRI coronagraphs, demonstrating their expected performance and potential for exoplanet characterization in the mid-infrared range.

Contribution

It provides the inaugural on-sky contrast data for MIRI's coronagraphs, including the first implementation of the four quadrant phase mask in space.

Findings

Raw contrasts >10^3 at <1'' separation

Contrast improves to ~10^5 beyond 5-6''

Reference star subtraction meets expectations

Abstract

Characterization of directly imaged exoplanets is one of the most eagerly anticipated science functions of the James Webb Space Telescope. MIRI, the mid-IR instrument has the capability to provide unique spatially resolved photometric data points in a spectral range never achieved so far for such objects. We aim to present the very first on-sky contrast measurements of the MIRI's coronagraphs. In addition to a classical Lyot coronagraph at the longest wavelength, this observing mode implements the concept of the four quadrant phase mask for the very first time in a space telescope. We observed single stars together with a series of reference stars to measure raw contrasts as they are delivered on the detector, as well as reference subtracted contrasts. MIRI's coronagraphs achieve raw contrasts greater than $10^3$ at the smallest angular separations (within $1''$) and about $10^5$ further out (beyond $5\sim6''$). Subtracting the residual diffracted light left unattenuated by the coronagraph has the potential to bring the final contrast down to the background and detector limited noise floor at most angular separations (a few times $10^4$ at less than $1''$). MIRI coronagraphs behave as expected from simulations. In particular the raw contrasts for all four coronagraphs are fully consistent with the diffractive model. Contrasts obtained with subtracting reference stars also meet expectations and are fully demonstrated for two four quadrant phase masks (F1065C and F1140C). The worst contrast, measured at F1550C, is very likely due to a variation of the phase aberrations at the primary mirror during the observations, and not an issue of the coronagraph itself. We did not perform reference star subtraction with the Lyot mask at F2300C, but we anticipate that it would bring the contrast down to the noise floor.