Atmospheric correction - Ground refletances

[Gonçalo Correia da Fonseca Rodrigues]

Hi,

I am using Py6S  to convert TOA radiance images (Sentinel-3) to surface reflectances. To represent of target refletances in a code 6SV, I used the 'GroundReflectance.HomogeneousLambertian GroundReflectance.LakeWater)'.

However after atmospheric correction I obtained very high reflectances comparing with in situ data, mainly in band 1 of OLCI.

I tested use an array of values characterizing the typical spectrum of reflectances measured in situ. After running the 6S, my output (s.outputs.atmos_corrected_reflectance_lambertian) it was the same.

I also tested using 'GroundReflectance.HomogeneousLambertian GroundReflectance.Sand )'  and the result hasn't changed either.

In outputs, for 'reflectance at satellite level', the value changes with LakeWater or Sand. It looks all right, with higher reflectances using Sand.

But in the end of the output file in 'atmospheric correction result' section, there are not changes. By changing other inputs (AOT, WV, Altitude of Target, AOT550 and others) the surface reflectances is changed. Only for this parameter (GroundReflectance), there are no changes.

I don't understand what could be happening. I already tested uninstall and reinstall py6S but I didn't solve the problem.

Many thanks in advance,

Gonçalo

[Robin Wilson]

Hi,

The GroundReflectance parameter is not taken into account when running in the atmospheric correction mode. This is because the aim of the atmospheric correction is to try and determine the ground reflectance - so setting it in this mode makes no sense. The underlying 6S model will ignore any ground reflectance parameterisation when running in atmospheric correction mode.

I suspect the reason for getting very high reflectances compared to in situ data is because the parameterisation of the atmosphere isn't quite matching the actual conditions. If you're experiencing the most disparity in Band 1 (Blue) then I suspect part of the issue is a misparameterisation of aerosol optical thickness or aerosol type, as this will predominantly affect the blue end of the spectrum.

Hope that helps,

Robin

--
You received this message because you are subscribed to the Google Groups "Py6S" group.
To unsubscribe from this group and stop receiving emails from it, send an email to py6s+uns...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/py6s/309d9fa7-286e-4c38-91f9-726c68a0231bn%40googlegroups.com.