Research Byte

Published in the RSAA Lunations
Vol1 Issue21 1–31 October 2021

The origin of the diffuse γ-ray background.

Since about the 60s we have been observing an unresolved background of γ-ray emission which could not be definitively associated with any obvious sources. An analysis of this background was obtained using NASA’s Fermi LAT, a dedicated space-borne γ-ray observatory, following several years of integrated observation. Its origin has so far only been weakly constrained due to the lack of detailed models for the emission mechanisms in possible source candidates.

Two of the most likely contributors are different forms of AGN (mostly blazars) and star-forming galaxies. In the former, the γ-ray emission is produced in the relativistic jets that emerge from the supermassive black holes at the centre of the galaxies, whereas in the latter it is produced mostly in collisions of cosmic rays with the ISM. So far, most estimates of the contribution of each source have been made by extrapolating locally calibrated luminosity functions to high redshift, assuming some spectral shape for each source class and summing their contributions. Our research on the other hand tackles the problem for star-forming galaxies using a bottom-up approach.

Cosmic rays (CRs) are relativistic particles, ions and electrons, that are accelerated in the shock waves of supernova remnants. Most CR ions lose energy in one of two ways, they either collide with the gas of the ISM, or they diffusively escape the galactic disk. The former is a well understood process in particle physics and results in the production of large numbers of neutral and charged pions, where the former decay into two γ-rays which makes up most of the observable γ-ray emission, and the latter produce secondary electrons and positrons, and a large number of neutrinos.

The problem that has so far thwarted many attempts to directly model the associated emission spectra is the lack of a detailed understanding how of CRs move through the ISM and how this translates to the fraction of CRs that actually die inside the galaxy, also known as the calorimetry fraction. However recent work done at the MSO (MK, RC+ 2020), has shone more light on the problem by considering how CRs self-generate turbulence in the magnetic field, known as the streaming instability, leading to self-confinement and how this turbulence is suppressed by collisions of ISM ions with the neutral phase. Balancing these processes gives us an energy dependant calorimetry fraction.

We proceeded by building and verifying a model to predict the γ-ray spectra of individual galaxies, requiring only a few basic parameters, e.g. the star-formation rate, half-light radius, stellar mass and redshift, that can be observed out to high redshift. We further assume standard and well-established model parameters such as the conversion of star-formation to supernova rate, the fraction of supernova energy that goes into CRs, the slope of the injection spectrum, the mean ionisation fraction, and the mean Alfvén Mach number of the ISM for the systems that dominate the emission.

Applying this model to large sample of ~22,000 galaxies in a Hubble deep field survey (CANDELS), correcting for cosmic variance using the observed star-formation history of the Universe, we can derive a predicted spectrum for the diffuse γ-ray background. It turns out that we obtain a very good match to the observed spectrum without fine-tuning any of the parameters.

We are currently refining some of the leptonic emission models, from the primary (SNR injected) and secondary (charged pion decay) cosmic ray electrons and positrons. Working out the leptonic emission is slightly more involved than for CR ions, because significantly more energy loss mechanisms of similar orders of magnitude are available. We hope that this work will allows us to gain a better understanding of what drives the observed FIR-radio correlation, which is surprisingly tight over many orders of magnitude.

Matt Roth, Mark Krumholz, Roland Crocker and Silvia Celli (Università La Sapienza and INFN)

This research was published in Nature https://www.nature.com/articles/s41586-021-03802-x

Matt Roth

_{Photo credit: : NASA/DOE/Fermi LAT Collaboration.}