High-resolution panchromatic spectral models of galaxies including photoionization and dust

Abstract

An updated version of the dust radiation‐transfer code sunrise, including models for star‐forming regions and a self‐consistent calculation of the spatially dependent dust and PAH emission, is presented. Given a hydrodynamic simulation of a galaxy, this model can calculate a realistic 2D ultraviolet–submillimetre spectral energy distribution of the galaxy, including emission lines from H ii regions, from any viewpoint. To model the emission from star‐forming regions, the mappingsiii photoionization code is used. The high wavelength resolution (∼1000 wavelengths) is made possible by the polychromatic Monte Carlo algorithm employed by sunrise. From the 2D spectral energy distributions (SEDs), images in any filter bands or integrated galaxy SEDs can be created. Using a suite of hydrodynamic simulations of disc galaxies, the output broad‐band images and SEDs are compared with observed galaxies from the multiwavelength SINGS and SLUGS galaxy surveys. Overall, the output SEDs show a good match with observed galaxies in colours ranging from far‐UV to submillimetre wavelengths. The only possible exception is the 160/850 μm colour, which the simulations underestimate by a factor of ∼5 compared to the SINGS sample. However, the simulations here agree with the SLUGS galaxies, which consistently have significantly larger amounts of cold dust than the SINGS galaxies. The sunrise model can be used to generate simulated observations of arbitrary hydrodynamic galaxy simulations. In this way, predictions of galaxy formation theories can be directly tested against observations of galaxies.

Massive close pairs measure rapid galaxy assembly in mergers at high redshift

“…Because Illustris resolves galaxy structures on spatial scales below ∼ 1 kpc at z 1, its mock survey fields contain sources appropriate for comparing with existing deep, high resolution, but narrow survey fields. In preliminary work, we conducted a simple proof-of-concept to combine our Thin lightcone geometries with the spectral synthesis code Sunrise (Jonsson 2006;Jonsson et al 2010), and we will release these initial synthetic images to the community. For example, Vogelsberger et al (2014a) used these to compare the Illustris to the HST Ultra Deep Field.…”

Predicting the stellar and non-equilibrium dust emission spectra of high-resolution simulated galaxies with dart-ray

“…Then they calculate the dust emission spectra from the average of the radiation field spectra in each bin. The use of these two parameters to construct the SED adaptive library allowed those authors to approximately consider both the intensity and spectral shape of the radiation field, unlike the simpler approach of Wood et al (2008) and Jonsson et al (2010).…”

“…This allows the tabulation of some quantities often used in the calculation. Wood et al (2008) and Jonsson et al (2010) use libraries of precomputed dust emission SEDs and assume that the dust emission is determined by the amplitude but not by the spectral shape of the heating radiation. Finally, Baes et al (2011), further developing an approach first proposed by Juvela & Padoan (2003), construct a dust emission SED adaptive library for a given RT calculation, which is derived by grouping the inferred radiation field spectra in bins of similar equilibrium dust temperatures and effective wavelengths (see their section 3.5.2).…”

“…The SEDs of the simulated galaxy have been used by several authors to elucidate the dependence of multiwavelength emission from galaxies on their intrinsic physical properties (e.g. Hayward et al 2012, 2013, Chakrabarti & Whitney 2009, Jonsson et al 2010, Domínguez-Tenreiro et al 2014. One advantage of performing radiative transfer calculations for numerically simulated galaxies, which has not been considered very much so far, is that one can also study the predicted observed properties of these galaxies on local scales.…”

Galaxies in the Illustris simulation as seen by the Sloan Digital Sky Survey - I: Bulge+disc decompositions, methods, and biases.

“…Using the starburst99 single-age stellar population synthesis (SPS) models (Leitherer et al 1999;Vázquez & Leitherer 2005; Leitherer et al 2010) an SED is assigned to each stellar particle based on its mass, metallicity, and age. The sunrise code (Jonsson 2006;Jonsson et al 2010) is used to map stellar light from each particle and generate a synthetic image. The dust absorption, scattering, or emission functionalities of the sunrise code are not used in the radiative transfer (see Section 2.2.3).…”