For over ten years, I was part of the research group headed by Prof. Maarten Baes, concentrating on the interstellar gas and dust contained in nearby and distant galaxies. The group uses a combination of numerical simulations and astrophysical observations across the electromagnetic spectrum, with a particular interest in far-infrared and submillimeter astronomy.



SKIRT is the name of the code I have been working on. It is an advanced 3D Monte Carlo radiative transfer code, used to study the effects of dust absorption and scattering in astrophysical objects such as galaxies, AGN tori, and planetary disks.

For more information, visit the SKIRT web site ( )


Theoretical models

SKIRT simulations can use theoretical models to approximate reality. For example, the stellar populations in an edge-on spiral galaxy (such as the Sombrero galaxy) could be represented by an exponential disk and a Sersic bulge, and the dust distribution by a (much flatter) exponential disk.

Upper picture: the Sombrero galaxy observed in the V-band (visible light); the false color indicates brightness. The central dark line is caused by dust absorption; it is called the "dust lane".

Lower picture: a SKIRT model of the Sombrero galaxy, from the same viewing angle, built by Ilse De Looze (part of our research group).

Simulated galaxies

These days SKIRT is often used to post-process the results of detailed hydrodynamical simulations of galaxies (conducted by other research teams). In this case SKIRT imports the simulated numerical model of the stars and dust in a galaxy, and it simulates the effects of the dust on the stellar radiation. In the end SKIRT produces a set of images and a spectral energy distribution (SED) for arbitrary viewing angles, showing us how this simulated galaxy would be observed if it were in the sky. Since the simulated geometries are more complex and have finer structure, the resulting images are more realistic as compared to those generated from theoretical models.

This picture is a SKIRT rendering of a simulated galaxy obtained from A. Rahimi.

Dust grid

To build a discretized approximation of the radiation field in a dusty medium, SKIRT partitions the spatial domain using a three-dimensional grid, assuming that all relevant physical quantities are constant within each grid cell. Needless to say, both the accuracy of the results and the simulation time scale with the total number of cells. An optimal grid provides high resolution where it is needed (i.e. in areas with a lot of dust, or with a high gradient in the radiation field) without placing too many cells in less demanding areas.

With Waad Saftly (then part of our research group) I investigated ways to construct optimal dust grids. To the left is a section through a grid for the simulated galaxy shown above (this particular grid looks nice but unfortunately is not really optimal).


EAGLE is a suite of hydrodynamical simulations (see the EAGLE home page) that follow the formation of galaxies in a cosmologically representative volume, using billions of particles to represent stars, gas and dark matter. The simulation code includes detailed recipes for star formation, gas cooling, stellar evolution, and feedback from supernovae and massive black holes. The simulation results reproduce many observed properties of present-day galaxies to unprecedented levels of agreement.

I selected a few hundred typical galaxies from the EAGLE simulation results and post-processed them with SKIRT to compare "mock observations" of the EAGLE galaxies with real-world observations of similar galaxies. Here are some SKIRT images of an EAGLE galaxy.

Above top row: visible light composites from two different viewing angles; the dark patches are caused by dust absorption.
Above bottom row: far-infrared false color composites showing emission from the dust plus some residual starlight.
Below: the spectral energy distribution (SED) of the simulated galaxy from a few different viewing angles.

galaxy sed

And finally, below is a SKIRT movie simulating a flight around and through the same EAGLE galaxy.

© Peter Camps 2013-2023