Mismatch of agnpy's absorption with Finke (2016) reference: DT about agnpy HOT 6 CLOSED

After removing the dependency on the Blob from the absorption (in PR #76), the case of the dust torus at all distances has improved. In the agnpy comparison now mu_s=1, before mu_s was taken from the blob.
Here some examples, the final check compares against a point source approximating the disk in case of a very large from the BH.

I would close this, if it's ok with you @jsitarek.
I think we have a decent agreement for this case now.

from agnpy.

the agreement looks very nice, closing

from agnpy.

sorry for reopening it, but I found something strange in DT absorption while investigating the BLR one.

if r << R_dt the reference line and the agnpy line are basicaly shifted by a factor very close to 1+z. This is something that we were discussing before, but from the paper it seems that the curves are given in energy at the galaxy, and actually for BLR absorption from the same plot there is a very nice agreeement.

if the emission region is far there starts to be a disagreement between agnpy and Finke's paper.
However I'm not sure of the actual parameters, table3 gives parameters for extended dust torus, while the plot is for point like. I used T=1000K (which is probably fine), and R_dt = 3.5e19 cm (which is the outer DT radius in the table for the extended DT).
@cosimoNigro which parameters have you used in your comparison plots?

from agnpy.

Following the discussion that we had today with @cosimoNigro I updated the plots with the proper R_dt= 1.6e19 cm (using formula 96 from the Finke'16 paper). If the Finke's points are scaled up by a factor of 2 (an issue in that paper that we discussed earlier), and are corrected by redshift of then I get a perfect agreement like Cosimo in his plots. We can close again the issue. However I think it is worth to write to Justin Finke that the BLR and DT absorption points in his Fig 14 seem to be computed w.r.t. different energies (BLR in the source frame, DT in the observer frame).

from agnpy.

@jsitarek can we close after #89?

from agnpy.

yes, closing...

from agnpy.