Table of Contents
Fetching ...

Mid-infrared extinction toward the Galactic center

Sebastiano D. von Fellenberg, Joseph M. Michail, S. P. Willner, Braden Seefeldt-Gail, Tamojeet Roychowdhury, Macarena Garcia Marin, Giovanni G. Fazio, Nicole M. Ford, Daryl Haggard, Joseph L. Hora, Howard A. Smith, Zach Sumners, Gunther Witzel

TL;DR

The paper tackles the long-standing challenge of mid-infrared extinction toward the Galactic center by integrating continuum dust-emission modeling with line-derived extinction measurements from JWST MIRI/MRS observations, anchored by new 15 GHz VLA free-free data. It introduces a two-dimensional Gaussian-field approach to recover the line-of-sight dust opacity distribution and provides a Python tool for deriving intrinsic dust spectra, yielding a refined MIR extinction law that varies spatially yet remains broadly consistent with prior Galactic-center work. The study finds spatial extinction variations, higher local extinction near dusty sources like IRS 29N, and no detectable PAH features, with a best-guess extinction curve that aligns with other galaxies while highlighting the dominant role of a single absorbing screen in the CMZ. These results substantially improve the accuracy of MIR extinction corrections for Sgr A* and nearby MIR sources, with quantified residual uncertainties that support future measurements of flares and Sgr A*’s spectral energy distribution.

Abstract

We determine the mid-infrared (MIR, $\sim$5~μm--22~μm) extinction towards the Galactic center using MIRI/MRS integral field unit (IFU) observations of the central $3''\times3''$ region (near 5~μm) to $7''\times7''$ region (near 22~μm). To measure the MIR extinction, we employ two approaches: modeling the intrinsic-to-observed dust thermal spectrum and assessing the differential extinction between hydrogen recombination lines. Expanding on prior work, we directly model the dust-opacity distribution along the line of sight, and we make available a Python code that provides a flexible tool for deriving intrinsic dust emission spectra. We confirm the spatial variability of extinction across the field, demonstrating that dusty sources--such as IRS~29N--exhibit higher local extinction. Furthermore, we verify the absence of PAH emission features in the Galactic center MIR spectra. Using the two complementary methods, we derive a refined ``best guess'' MIR extinction law for Sgr A* and the surrounding Galactic-center region. By applying the extinction law to a MIR flare measurement discussed in a companion paper Michail et al. 2025, we estimate a residual relative extinction uncertainty for the short MIRI/MRS grating on the order of $0.2~\mathrm{mag}$ {from $\sim$5 to $\sim$18~μm\ and $\sim$0.3~mag from $\sim$18 to $\sim$22~μm}, consistent with our uncertainty estimate.

Mid-infrared extinction toward the Galactic center

TL;DR

The paper tackles the long-standing challenge of mid-infrared extinction toward the Galactic center by integrating continuum dust-emission modeling with line-derived extinction measurements from JWST MIRI/MRS observations, anchored by new 15 GHz VLA free-free data. It introduces a two-dimensional Gaussian-field approach to recover the line-of-sight dust opacity distribution and provides a Python tool for deriving intrinsic dust spectra, yielding a refined MIR extinction law that varies spatially yet remains broadly consistent with prior Galactic-center work. The study finds spatial extinction variations, higher local extinction near dusty sources like IRS 29N, and no detectable PAH features, with a best-guess extinction curve that aligns with other galaxies while highlighting the dominant role of a single absorbing screen in the CMZ. These results substantially improve the accuracy of MIR extinction corrections for Sgr A* and nearby MIR sources, with quantified residual uncertainties that support future measurements of flares and Sgr A*’s spectral energy distribution.

Abstract

We determine the mid-infrared (MIR, 5~μm--22~μm) extinction towards the Galactic center using MIRI/MRS integral field unit (IFU) observations of the central region (near 5~μm) to region (near 22~μm). To measure the MIR extinction, we employ two approaches: modeling the intrinsic-to-observed dust thermal spectrum and assessing the differential extinction between hydrogen recombination lines. Expanding on prior work, we directly model the dust-opacity distribution along the line of sight, and we make available a Python code that provides a flexible tool for deriving intrinsic dust emission spectra. We confirm the spatial variability of extinction across the field, demonstrating that dusty sources--such as IRS~29N--exhibit higher local extinction. Furthermore, we verify the absence of PAH emission features in the Galactic center MIR spectra. Using the two complementary methods, we derive a refined ``best guess'' MIR extinction law for Sgr A* and the surrounding Galactic-center region. By applying the extinction law to a MIR flare measurement discussed in a companion paper Michail et al. 2025, we estimate a residual relative extinction uncertainty for the short MIRI/MRS grating on the order of {from 5 to 18~μm\ and 0.3~mag from 18 to 22~μm}, consistent with our uncertainty estimate.

Paper Structure

This paper contains 15 sections, 2 equations, 11 figures, 2 tables.

Figures (11)

  • Figure 1: 15 GHz VLA image. The grey box indicates the region used to model the Mini-spiral extinction, and the location of Sgr A* is indicated with a white star. The white circle in the bottom left indicates the beam size smoothed to JWST/MIRI 17.98 $\mu$m resolution.
  • Figure 2: Continuum model (top left) and derived extinction law (top right) derived from a fit to the spectrum in the Sgr A* pixel. The black line and dots show the binned spectrum with an artificially added $3\%$ log-normally distributed uncertainty. The grey lines show posterior samples of the extincted dust model. The panel below shows the $\chi^2$ residual. The right plot shows the derived extinction curves ($A_\lambda$). The orange line shows the median posterior model, and the black line shows the posterior samples. The dashed-grey line shows the extinction law derived by Fritz2011. That law is based on the ISO/SWS observations by Lutz1996, which had an effective beam size of $14"\times16"$.
  • Figure 3: Dust distribution functions $\Psi(T, \tau_{9.8})$ for the Sgr A* pixel (left), the Mini-spiral (center), and IRS 29 (right). The insets in the Sgr A* and Mini-spiral panels show zoom-ins on the peaks of the respective posterior distributions.
  • Figure 4: Dust continuum models and derived extinction laws for different regions observed with JWST MIRI/MRS. The left panel shows the continuum fits to apertures centered on Sgr A*, the Mini-spiral, and the MIR-bright star IRS 29N. The right panel shows the corresponding extinction laws. The grey-dashed line shows the MIR-extinction law derived by Fritz2011.
  • Figure 5: Posterior distribution of the optical depth law $\tau(\lambda)$ defined in \ref{['eq:dust_continuum']}. Different colors show the optical depths for different regions. The dashed line shows the optical depth profile by Kemper2004, which serves as a prior model and from which we allowed a multiplicative deviation.
  • ...and 6 more figures