Dear Patrick,
I am not sure to understand very well but I am running ARTS with the atmlab
package and I am wondering if your changes might affect my simulations.
In my controle file I declare Q.BLACKBODY_RADIATION_AGENDA = {
'blackbody_radiation_agenda__Planck' };
and then I define Q.IY_UNIT='PlanckBT' to output brightness temperatures
instead of radiances.
Are your modifications in line with these structures ? If yes, does it mean
that the output
from the radiative transfer will be radiances and we will have to convert these
radiances into brightness temperatures ?
Thank you.
Best regards,
Pauline
----- Météo-France -----
Dr. Pauline Martinet
Chercheur CNRM/GMEI/LISA
[email protected]
Fixe : +33 561079031
Site web: www.sites.google.com/site/martinetpauline31
----- Mail original -----
De: "Patrick Eriksson" <[email protected]>
À: [email protected], [email protected], [email protected]
Envoyé: Vendredi 11 Décembre 2015 15:37:43
Objet: [Qpack] Goodbye to blackbody_radiation_agenda?
Hi all,
Is anyone changing blackbody_radiation_agenda from its default?
In line with ARTS philosophy of modularity, I once introduced
blackbody_radiation_agenda. The default setting for this agenda is the
Planck function, defined with respect to frequency. You can also define
the Planck function with respect to wavelength. Or apply the
Rayleigh-Jeans approximation and set blackbody_radiation to the
temperature. The question is if anyone is using these latter options?
And if yes, would it be critical to remove the agenda and hard-code what
is now the default? It would still be possible to convert final
radiances to other units, including wavelength-based ones.
This issue has already created problems around DOIT. It makes the
conversion to other units less secure. Richard assumes in the non-LTE
part that the default is valid. Richard is now extending the scope of
the analytical calculation of Jacobians, and this would be much easier
with a hard-coded solution. It could be motivated to add methods to
mimic details of the calibration process, but this would be complicated
with arbitrary source function.
In addition, is it possible to simple change the source function without
adopting other things? For example, is the line shape function dependent
on the source function used? I am here mainly thinking about the (v/v0)
pre-factor. Could the different (linear or quadratic in v/v0) here be
associated with this issue? Some other theoretical consideration here?
In summary, Richard and I want to remove the agenda, both for practical
and theoretical reasons. But we don't want to cause problems for
someone. If I don't hear anything I maybe start doing changes already
Tuesday afternoon.
Bye,
Patrick
_______________________________________________
qpack mailing list
[email protected]
https://www.sat.ltu.se/mailman/listinfo/qpack
_______________________________________________
qpack mailing list
[email protected]
https://www.sat.ltu.se/mailman/listinfo/qpack
