A radially broad collisional cascade in the debris disk of $γ$ Ophiuchi observed by JWST

TL;DR

JWST/MIRI imaging of gamma Oph reveals a radially broad, smooth debris disk extending to ~250 au at 25.5 μm, consistent with a broad planetesimal belt undergoing a steady-state collisional cascade with a single grain-size distribution. The disk exhibits a mild stellocentric offset that can be explained by a perturbing planet outside ~10 au (up to ~10 M_Jup) or a stellar-mass companion at a few AU, without producing a resolvable gap. Across wavelengths, the inner warm dust is modest and not clearly PR-drag–driven, and ALMA data support a collision-dominated, extended belt. Collectively, gamma Oph emerges as a rare archetype of a radially broad, collision-dominated debris disk, in contrast to Vega and Fomalhaut, and provides tight constraints on the planetary architecture that sculpt such systems.

Abstract

The A1V star $γ$ Oph, at a distance of 29.7 pc, is known from Spitzer imaging to host a debris disk with a large radial extent and from its spectral energy distribution to host inner warm dust. We imaged $γ$ Oph with JWST/MIRI at 15 and 25.5 microns, which reveal smooth and radially broad emission that extends to a radius of at least 250 au at 25.5 microns. In contrast to JWST findings of an inner small-grain component with distinct ringed substructures in Fomalhaut and Vega, the mid-infrared radial profile combined with prior ALMA imaging suggests a radially broad steady-state collisional cascade with the same grain size distribution throughout the disk. This further suggests that the system is populated by a radially broad planetesimal belt from tens of au or less to well over 200 au, rather than a narrow planetesimal belt from which the observed dust is displaced to appear broad. The disk is also found to be asymmetric, which could be modelled by a stellocentric offset corresponding to a small eccentricity of $\sim$0.02. Such a disk eccentricity could be induced by a mildly eccentric $<$$10\,M_\mathrm{Jup}$ giant planet outside 10 au, or a more eccentric companion up to stellar mass at a few au, without producing a resolvable radial gap in the disk.

Figures (10)

• Figure 1: PSF-subtracted MIRI F2550W (top row) and F1500W (bottom row) images. The left column subtracts off a point source centred on the star scaled to the stellar flux density inferred from the SED (205 mJy for F2550W and 591 mJy for F1500W; see Section \ref{['sec:almased']} and Fig. \ref{['fig:sed']}), in which the central component remains bright. The right column subtracts off a higher PSF flux density (246 mJy for F2550W and 615 mJy for F1500W) to further emphasise the structure of the resolved disk component. PSF subtraction was performed dither by dither with stage 2 products before combining the subtracted dithers with the stage 3 pipeline to produce the images displayed here. The images are displayed on a logarithmic scale. The PSF FWHM is denoted by the shaded ellipse in the bottom left corner of each panel. The orientation of the images is North up, East left.
• Figure 2: Gallery of nonparametric and parametric models fitted to the MIRI F2550W PSF-subtracted (assuming stellar SED flux) image of $\gamma$ Oph. Each group of 3 panels displays the PSF-convolved disk-only model image (i.e., stellar component not included, displayed on a logarithmic colour scale, to be compared with the top-left panel in Fig. \ref{['fig:obs']}), disk-only deconvolved and deprojected radial surface brightness (S.B.) and optical depth ($\tau$) profiles and the residual image (data $-$ model). "Offset" indicates that the geometric centre of the disk is offset relative to the star in the model, and is included as (two) free parameters in the model (along two spatial axes). The displacement vector of the disk centre relative to the star is indicated with blue arrows. All radial profile panels are plotted with the same vertical scaling, which span from 0 to 1.8 mJy/arcsec$^2$. For the rave radial profile panel (panel a), the azimuthally averaged profile of the PSF-subtracted (assuming the stellar flux of the best-fit grid point in the 4D rave grid) observations and the PSF-convolved best-fit disk model are overplotted, as this is the 1D quantity that rave directly fits to. Where multiple Gaussian components are invoked (panels e and f), the constituent Gaussian components are individually overplotted. Uncertainties are indicated with shaded regions for the main deconvolved profiles, but these are generally too narrow to be visible for the parametric models. The per-pixel RMS noise measured from the background is 0.07 mJy/arcsec$^2$. North is oriented upwards and east to the left.
• Figure 3: Assumed parameters for rave and fitted values for parametric models at 25 $\mu$m.
• Figure 4: Gaussian model fitted to the MIRI F1500W image of $\gamma$ Oph. The panels are the same as those described in Fig. \ref{['fig:models25']} for the F2550W modelling and the model image should be compared with the bottom-right panel of Fig. \ref{['fig:obs']}. The central region has been masked as it is dominated by PSF subtraction artefacts and was not used for model fitting. The per-pixel RMS noise measured from the image background is 0.02 mJy/arcsec$^2$.
• Figure 5: ALMA Band 7 observations presented in Marino2025, imaged with clean using a robust parameter of 2.0 and with primary beam correction applied. The beam FWHM is indicated with a white ellipse in the bottom-left corner. Contours are drawn at 0.06, 0.12 and 0.15 mJy arcsec$^{-2}$ based on the image smoothed with a 1 arcsec UV taper.