You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When computing the absorption spectrum of water in the 2725-2732 nm range, spectreatm.py displays a jump at 2729.5 nm
The problem does not occur with a custom-made Matlab program reading .par files and computing Voigt profiles
I have checked that the Hitran lines download with HAPI have a 100% match with the ones I am using in my other program
Increasing eps or the resolution does not solve the problem
This is not due to missing line data, because it works with my other program and hey, we're dealing with the main isotope of water here
I think it may have something to do with the Voigt code not taking into account enough neighboring lines, i.e. as if the default values for the 'OmegaWing' and/or 'IntensityThreshold' were incorrect.
We were on the verge of migrating our matlab code to Python but it gave m a cold chill.
Please tell me if you can reproduce the issue. Thanks.
this is because the default value for DefaultOmegaWingHW = 50 is too low
setting it at 100 fixes the problem
So here's what I suggest:
change 50 to 100 and update HAPI, OR
create a custom parameter in compute_absorption() that will be passed to absorptionCoefficient_Voigt, that will be passed to absorptionCoefficient_Generic, so that OmegaWingHW=DefaultOmegaWingHW is overriden by a user-provided value, or multiplied by a user-provided factor (no unit)
Hi,
When computing the absorption spectrum of water in the 2725-2732 nm range, spectreatm.py displays a jump at 2729.5 nm
I think it may have something to do with the Voigt code not taking into account enough neighboring lines, i.e. as if the default values for the 'OmegaWing' and/or 'IntensityThreshold' were incorrect.
We were on the verge of migrating our matlab code to Python but it gave m a cold chill.
Please tell me if you can reproduce the issue. Thanks.
My code is the following:
My colleague told me that the error happens within this piece of code:
spectreatm / HAPI result:
Matlab (in-house code) result:
The text was updated successfully, but these errors were encountered: