search

GEUS-SICE / pySICE

Python and Fortran scripts behind the SICE toolchain for albedo retrieval.
GNU General Public License v2.0
5 stars 1 forks source link

Update backscattering coefficient #6

Open BaptisteVandecrux opened 3 years ago

BaptisteVandecrux commented 3 years ago

From: Alex Kokhanovsky Sent: 7. januar 2021 09:44 To: Baptiste Vandecrux Subject: Re: Asymmetry parameter and backscatter fraction

please, use the formulation as in our 2020 paper

On Wed, 6 Jan 2021 at 08:30, Baptiste Vandecrux wrote: Thank you Alex. I see in the fortran code that you calculate the atmospheric transmittance as:

     wa1=1.10363
     wa2=-6.70122

     wx0=2.19777
     wdx=0.51656

     bex=exp   (  (g-wx0)/wdx )
     sssss=  (wa1-wa2)/(1.+bex)+wa2
     t1=exp(-(1.-g)*tau/am1/2./sssss

Since t1 is defined in Eq. A9 and A10 of Kokhanovsky et al. (2020) as:

    t1=exp(-Batm*tau/am1)

where Batm is the atmospheric backscattering coefficient. It means that your formulation implies:

    Batm = (1.-g)/2./sssss

Surprisingly, this approximation of Batm do not match with the definition of Batm given in Kokhanovsky et al. 2020:

image

The difference is striking:

image

Could you check whether the approximation is accurate?

Best, Baptiste