Investigating the cause of the $\alpha\,$-$\,z$ relation

The correlation between radio spectral index and redshift has long been used to identify high redshift radio galaxies, but its cause is unknown. Traditional explanations invoke either $(i)$ intrinsic relations between spectral index and power, $(ii)$ environmental differences at high redshift, or $(iii)$ higher inverse Compton losses due to the increased photon energy density of the cosmic microwave background. In this paper we investigate whether the increased inverse Compton losses alone can cause the observed spectral index - redshift correlation by using spectral modelling of nearby radio galaxies to simulate high redshift equivalents. We then apply selection effects and directly compare the simulated radio galaxy sample with an observed sample with sufficient redshift coverage. We find excellent agreement between the two, implying that inverse Compton losses and selection effects alone can largely reproduce the observed spectral index - redshift correlation.