Further correction to the STARFORGE methods paper: Planck-mean dust opacities
Michael Y. Grudić, Dávid Guszejnov, Philip F. Hopkins, Stella S. R. Offner, Claude-André Faucher-Giguère
TL;DR
This erratum addresses the unphysical divergence of the previous Planck-mean opacity $κ_P(T_d,T_{rad})$ at low $T_{rad}$ by replacing the limited-range fit with an updated calculation and explicit $T_{rad}$ dependence. It clarifies that the dust opacity depends on both $T_{rad}$ and $T_d$ out of LTE and derives a consistent energy-balance framework in which absorption and emission use the appropriate temperature arguments. The authors compute $κ_P(T_d,T_{rad})$ and the Rosseland mean opacity from semenov_2003 tables across five $T_d$ regions, providing a practical on-the-fly interpolant and publicly releasing the supporting code and tables. This work improves the robustness and generality of dust radiative transfer in STARFORGE and related simulations, enabling more accurate thermal evolution and fragmentation predictions in star-forming environments.
Abstract
The model for the Planck-mean dust opacity $κ_{P}$ given in Appendix C of the STARFORGE simulations methods paper does not extrapolate well to low radiation temperature $T_{\rm rad}$, so we provide an updated calculation suitable for general use. We also clarify the role of the dust and radiation temperatures in setting the dust opacity, and provide code and calculations of the Planck- and Rosseland- mean dust opacity as a function of both the dust temperature $T_{\rm d}$ and the radiation temperature $T_{\rm rad}$.
