Synthesizer: a Software Package for Synthetic Astronomical Observables

TL;DR

Synthesizer tackles the challenge of rapidly generating consistent, multi-wavelength observables from galaxy simulations by delivering a modular, extensible framework that unifies stellar, gas, and AGN emission with dust, radiative transfer approximations, and instrument effects. Its core design centers on Galaxy objects, Grid-based emission models, and a flexible EmissionModel pipeline that produces spectra, lines, photometry, and imaging, all orchestrated via a Pipeline for scalable, parallel execution. The work provides a practical platform for forward modelling, enabling robust exploration of modelling choices and large training-set generation for SBI, while complementing high-fidelity RT approaches. Its open-source nature, comprehensive testing, and emphasis on speed and interoperability position Synthesizer as a versatile tool for forward and inverse modelling across cosmological simulations and observational pipelines.

Abstract

We present Synthesizer, a fast, flexible, modular and extensible platform for modelling synthetic astrophysical observables. Synthesizer can be used for a number of applications, but is predominantly designed for generating mock observables from analytical and numerical galaxy formation simulations. These use cases include (but are not limited to) analytical modelling of the star formation and metal enrichment histories of galaxies, the creation of mock images and integral field unit observations from particle based simulations, detailed photoionisation modelling of the central regions of active galactic nuclei, and spectro-photometric fitting. We provide a number of stellar population synthesis models, photoionisation code configurations, dust models, and imaging configurations that can be used 'out-of-the-box' interactively. The code can be used to quantitatively test the dependence of forward modelled observables on various model and parameter choices, and rapidly explore large parameter ranges for calibration and inference tasks. We invite and encourage the community to use, test and develop the code, and hope that the foundation developed will provide a flexible framework for a number of tasks in forward and inverse modelling of astrophysical observables. The code is publicly available at https://synthesizer-project.github.io/

Figures (17)

• Figure 1: Schematic showing the main modules available in Synthesizer and how they relate to each other, described in Section \ref{['sec:design']}. Most forward modelling applications will combine a Grid object, a Galaxy object (consisting of stars, gas and black holes components) and an Emission Model object, and produce a number of observables (Sed's, images, etc.).
• Figure 2: An example showing some of the forward modelled observables that can be produced using Synthesizer. We show a galaxy taken from the Illustris TNG50 simulation, processed using a line of sight dust attenuation model, assuming a fixed dust to metal ratio, and using the BPASS v2.2.1 SPS model, with binary stars and a Kroupa IMF. Top row: full colour RGB images in, from left to right, JWST/NIRCam [F115W,F150W,F200W], HST/WFC3 [F450W,F547M,F814W], and Euclid NISP [YJH], where each band is linearly combined. The NIRCam image is produced using a PSF taken from perrin_updated_2014. Middle row, left to right: stellar mass map, mass-normalised metallicity map, mass-normalised age map, optical depth map (obtained from the gas particles), H-$\alpha$ emission map. Bottom row, left to right: the star formation - metal enrichment history (with marginal age and metallicity distributions), intrinsic and dust attenuated spectra and derived photometry (for edge on and face on orientations), histogram of optical depth for face-on and edge-on line of sight calculations, BPT diagram (star shows the integrated ratios, points show individual star particles, lines show the kauffmann_unified_2000kewley_theoretical_2013 classification regions).
• Figure 3: An idealised ROGB image of the same galaxy as in Figure \ref{['fig:mosaic']} in the JWST NIRCam F444W, F277W, F200W, F150W (left) and HST WFPC2 F814W, F606W, F547M, F450W (right) filters, using the make_lupton_rgb scheme in astropy using custom weightings. The top panel shows the face-on view while the bottom panel shows the edge-on view. This utilises a higher angular resolution compared to the native resolution of JWST and HST.
• Figure 4: Example of defining parametric and particle star formation and metal enrichment histories (SFZH). Left panel: randomly initialised star particles with a range of ages and metallicities, and their marginal distributions. Right panel: three parametric star formation history forms provided in Synthesizer, a constant, exponential and log-normal star formation history (SFH), each at fixed metallicity. Middle panel: The composite star formation and metal enrichment history (SFZH) from the combination of all the above particle and parametric objects. This functionality allows the generation of arbitrarily complex SFZH distributions.
• Figure 5: Example of the spectra contained within a grid object. Top left: all the different spectra produced from a grid, with age 5 Myr and metallicity 0.01, using the BPASS v2.2.1 SPS binary model with a Kroupa IMF, and post-processed with Cloudy with the fiducial photoionisation parameters. Top right: multiple incident grids assuming different SPS models (BC03, BPASS, FSPS, M24) all at the same age (5 Myr) and metallicity ($Z = 0.01$) Bottom left: variation of the same model in the top left panel with age, at fixed metallicity ($Z = 0.01$). Bottom right: variation of the same model in the top left panel with metallicity, at fixed age (5 Myr).