The complex inner disk of the Herbig Ae star HD 100453 with VLTI/MATISSE

TL;DR

The paper extends infrared interferometry of HD 100453 into the L-band with VLTI/MATISSE and combines these data with archival H- and K-band results to characterize a misaligned inner disk at sub-au scales. Through parametric modelling, time-variability analysis, and image reconstruction, the study finds an inner disk inclined at about $\sim 47.5^\circ$ with a position angle near $\sim 83.6^\circ$, consistent with significant inner-outer disk misalignment, and an inner radius of $R_{in} \approx 0.27$ au. The L-band data require higher-order, chromatic asymmetries—potentially due to dynamic interactions or disk instabilities—and a two-component morphology is favored by image reconstruction, with a temperature-gradient model jointly fitting MATISSE, GRAVITY, and PIONIER. These results highlight a dynamically complex inner disk whose true nature awaits confirmation from simultaneous, multi-wavelength interferometric campaigns across short- and long-baseline facilities. The findings have important implications for understanding inner-disk physics and planet-formation environments in misaligned systems.

Abstract

The inner regions of planet-forming disks hold invaluable insights for our understanding of planet formation. The disk around the Herbig star HD 100453 presents one such environment, with an inner disk that is significantly misaligned with respect to the outer disk. This paper expands the existing H-band (PIONIER) and K-band (GRAVITY) interferometric studies of the HD 100453 inner disk to the L-band with the MATISSE VLTI instrument. With snapshot data spanning approximately four years we aim for a more comprehensive understanding of the inner disk structures and their potential time evolution. Based on the MATISSE data obtained, we use a combination of analytical models and image reconstruction to constrain the disk structure. Additionally, we fit a temperature gradient model to the selected wavelength range of PIONIER, GRAVITY and MATISSE to derive physical properties of the inner regions. Our parametric model finds an inclination of $\sim 47.5^\circ$ and a position angle of $\sim 83.6^\circ$, which corroborates the case of strong inner-outer disk misalignment. From the symmetric temperature gradient we derive an inner disk radius around $\sim0.27$ au, with dust surface densities of $Σ_{\rm{subl}} \approx 10^{-3.2}$ g/cm$^2$ and vertical optical depth $τ_{\rm{z, subl}} \approx 0.1-0.06$. Same-night MATISSE and GRAVITY observations indicate the necessity for higher-order asymmetries to explain the interferometric signals, which is further supported by a MATISSE snapshot image reconstruction. The chromatic interferometric data reveal the likely need for higher-order asymmetries to explain the inner disk of HD~100453, suggesting a possible origin in dynamic interactions or disk instabilities. Coordinated multi-wavelength infrared interferometric observations with GRAVITY and MATISSE will be crucial to confirm these findings and uncover its underlying nature.

Figures (12)

• Figure 1: Overview of $uv$ space and timeline covered by the selected MATISSE observations. Each night is indicated with a unique marker. The colour represents the AT station configurations.
• Figure 2: Calibrated squared visibilities and closure phases from all the selected MATISSE observations.
• Figure 3: Fit of the MATISSE data following a similar time-invariant parametric model and methodology as 2017AA...599A..85L and 2022AA...658A.183B (Model-2, see Table \ref{['tab:model_fit_onlymatisse']}). Panels A and B show the observations and the best-fit model simulated squared visibilities and closure phases. Panel C shows the model brightness distribution $I_\nu\rm{(fit)}$, with panel D highlighting the skewed asymmetry following equation \ref{['eq:highlight_asymm']}.
• Figure 4: Violin-plot showcasing the time-variable MCMC fitting results of the asymmetry angle ($\theta_{\rm{skwPA}}$) for six epochs of MATISSE data, where the other disk parameters are fixed to the best-fit model-two asymmetric disk parameters from Table \ref{['tab:model_fit_onlymatisse']}. The horizontal dashed line shows the best fitted value for the asymmetry angle for the time-invariant dataset. Above each violin the change in the respective $\chi^2_r$ is shown compared to the time-invariant value. The error bars indicate the best fit value with 16/84% percentile intervals.
• Figure 5: Panel A shows the L-band image reconstruction for the 2021 MATISSE AT observations, which contain the large-medium-small arrays observed within 11 days, at $\lambda = 3.6 \pm 0.2~\mu$m using the SPARCO/MIRA software. The central star with $T=7250$ K, modelled as a point source centred on (0,0), is not depicted. Over-plotted are ellipses where the semi-major axis are the half-light radius $l_a$, taken from the best parametric fits of Table \ref{['tab:model_fit_onlymatisse']}. Panel B shows the approximate dirty beam of the combined observations at $\lambda = 3.6~\mu$m, together with a small inset of the $uv$ points.