Dear Patrick,

Thank you for this clarification.

I will see if I manage to remove the differences with your suggestions.
But as you think it is not important, I will not spend too much time on it.

We are comparing ARTS with another radiative transfer model and I was used to 
simulate
my spectra with Qpack. But then I used arts to simulate monochromatic pencil 
beam spectra as suggested
by Ole Martin.
And as I saw differences between the two calls I was a little bit worried. For 
me, ARTS is a little bit tricky 
and I did not want to have differences between my two radiative transfer models 
due to a bad use of ARTS.

But I will test your suggestions and see what happens.

Thank you again.

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: "Pauline Martinet" <[email protected]>
Cc: [email protected]
Envoyé: Lundi 7 Septembre 2015 20:53:30
Objet: Re: [Qpack] differences between Qpack and arts simulation

Pauline,

Let me first say that I don't think the differences you find are of any 
practical consequences, so probably not an "error". But I like that you 
are so careful.

First I must ask where you found that qpack2 should mimic arts_y 
exactly? This is basically, true but there some subtle differences 
around HSE (and that qpack uses automatically the ATMDATA fields).


> - T field identical to the first dimension of Q.T.ATMDATA.DATA

Is Q.T.ATMDATA.DATA then just a vector? Otherwise this is not a 
sufficient criterion. If you want to check, this is what qpack does to 
set Q.T_FIELD:

Q.T_FIELD   = qarts_atm_field( Q, 't', mjd, ho );

where mjd and ho are set based on data information in Y.


> - VMR field identical to the ABS_SPECIES.ATMDATA.DATA : ok

Same comment here.


> - Y.HSE_P equals to Q.HSE_P

This part is more tricky. First of all, if you have HSE.ON=true, Y.HSE_P 
and Y.HSE_Z will result in an altitude profile, that overwries anything 
you have in Q.Z_FIELD and Q.Z.ATMDATA. I will make this clear in 
qpack2.pdf (for v2.3).

With HSE.ON = true we have in fact a difference between running qpack 
and arts_y. To mimic exactly what qpack is doing you need to do

Q.Z_FIELD   = qarts_hse( Q, Y.HSE_P, Y.HSE_Z );

As you noticed, there is no Q.HSE.Z. When running outside qpack2, 
"HSE_Z" is taken from Z_FIELD by interpolation to HSE_P. And you have 
likely a small difference between Q.Z_FIELD and the Z_FIELD produced by 
qarts_hse.

If you set Q.Z_FIELD and set Q.HSE.ON=false (and don't set Z.ATMDATA), 
you should use identical altitudes both when running qpack2 and arts_y.

I hope this clarified things around HSE!?


> - Y.ZA is equal to Q.SENSOR_LOS (for the zenith angle)
> - Q.SENSOR_POS equals Y.Z_PLATFORM

OK

Bye,

Patrick
_______________________________________________
qpack mailing list
[email protected]
https://www.sat.ltu.se/mailman/listinfo/qpack

Reply via email to