There was an error in how we model extragalactic point sources in tools21cm.
We now create the 2D distribution of the sources at the first redshift bin and then calculate their frequency evolution with an exponential. In the end, you have the same amount of sources at the same position for every redshift bin. They are correlated in the frequency direction, similarly to the synchrotron foreground but just for one pixel.
There was an error in how we model extragalactic point sources in tools21cm. We now create the 2D distribution of the sources at the first redshift bin and then calculate their frequency evolution with an exponential. In the end, you have the same amount of sources at the same position for every redshift bin. They are correlated in the frequency direction, similarly to the synchrotron foreground but just for one pixel.