Search

Aims. We introduce a new deep-learning approach for the reconstruction of 3D dust density and temperature distributions from multi-wavelength dust emission observations on the scale of individual star-forming cloud cores (<0.2 pc). Methods. We constructed a training data set by processing cloud cores from the Cloud Factory simulations with the POLARIS radiative transfer code to produce synthetic dust emission observations at 23 wavelengths between 12 and 1300 µm. We simplified the task by reconstructing the cloud structure along individual lines of sight (LoSs) and trained a conditional invertible neural network (cINN) for this purpose. The cINN belongs to the group of normalising flow methods and it is able to predict full posterior distributions for the target dust properties. We tested different cINN setups, ranging from a scenario that includes all 23 wavelengths down to a more realistically limited case with observations at only seven wavelengths. We evaluated the predictive performance of these models on synthetic test data. Results. We report an excellent reconstruction performance for the 23-wavelength cINN model, achieving median absolute relative errors of about 1.8% in log(n/m−3) and 1% in log(Tdust/K), respectively. We identify trends towards an overestimation at the low end of the density range and towards an underestimation at the high end of both the density and temperature values, which may be related to a bias in the training data. After limiting our coverage to a combination of only seven wavelengths, we still find a satisfactory performance with average absolute relative errors of about 2.8% and 1.7% in log(n/m−3) and log(Tdust/K). Conclusions. This proof-of-concept study shows that the cINN-based approach for 3D reconstruction of dust density and temperature is very promising and it is even compatible with a more realistically constrained wavelength coverage. [ABSTRACT FROM AUTHOR]

Context. Mrk 71/NGC 2366 is the closest green pea (GP) analog and candidate Lyman Continuum (LyC) emitter. Recently, 11 LyC-leaking GPs have been detected through direct observations of the ionizing continuum, making this the most abundant class of confirmed LyC-emitters at any redshift. High resolution, multiwavelength studies of GPs can lead to an understanding of the method(s), through which LyC escapes from these galaxies. Aims. The proximity of Mrk 71/NCG 2366 offers unprecedented detail on the inner workings of a GP analog, and enables us to identify the mechanisms of LyC escape. Methods. We used 5825–7650 Å integral field unit PMAS observations to study the kinematics and physical conditions in Mrk 71. An electron density map was obtained from the [S II] ratio. A fortuitous second order contamination by the [O II] λ3727 doublet enabled the construction of an electron temperature map. Resolved maps of sound speed, thermal broadening, "true" velocity dispersion, and Mach number were obtained and compared to the high resolution magnetohydrodynamic SImulating the LifeCycle of molecular Clouds (SILCC) simulations. Results. Two regions of increased velocity dispersion indicative of outflows are detected to the north and south of the super star cluster, knot B, with redshifted and blueshifted velocities, respectively. We confirm the presence of a faint broad kinematical component, which is seemingly decoupled from the outflow regions, and is fainter and narrower than previously reported in the literature. Within uncertainties, the low- and high-ionization gasses move together. Outside of the core of Mrk 71, an increase in Mach numbers is detected, implying a decrease in gas density. Simulations suggest this drop in density can be as high as ∼4 dex, down to almost optically thin levels, which would imply a nonzero LyC escape fraction along the outflows even when assuming all of the detected H I gas is located in front of Mrk 71 in the line of sight. Conclusions. Our results strongly indicate that kinematical feedback is an important ingredient for LyC leakage in GPs. [ABSTRACT FROM AUTHOR]