Significantly different with Seadas different version to process Sentinel-2 data

[wxfei19]

Hello,

I used SEADAS l2gen to perform preliminary atmospheric correction on Sentinel-2 L1C data for the Yangtze River in China.
When mosaicking the data, I noticed noticeable color differences.
Upon inspection, I found that the rhos reflectance values obtained from different versions of SEADAS (9.7V 2024.5 and 9.11V 2025.5) showed significant discrepancies.

I am not very familiar with SEADAS and am unsure whether SEADAS has undergone substantial changes that led to such results. Personally, I believe the water body spectra obtained from version 9.11V 2025.5 appear relatively more reasonable.

the spectral profile files are attached.

I am sorry, I find the attached file could not show in this post.

the input data I used is S2A_MSIL1C_20160917T025542_N0500_R032_T50RKU_20230929T184803.SAFE downloaded from https://dataspace.copernicus.eu/

That's just unbelievable. I consider myself not very familiar with SeaDAS and haven't delved deeply into setting the parameters for the SeaDAS L2gen command.

But I ended up with significantly different data.


I used ncdump.exe to get the head information, the difference between them is the software_version "9.11.0-V2025.5" and "9.7.0-V2024.5" respectively.



// group attributes:
:software_name = "l2gen" ;
:software_version = "9.11.0-V2025.5" ;
:input_sources = "manifest.safe,none,GMAO_MERRA2.20160917T020000.MET.nc,GMAO_MERRA2.20160917T030000.MET.nc,GMAO_MERRA2.20160917T040000.MET.nc,GMAO_MERRA2.20160917T020000.MET.nc,GMAO_MERRA2.20160917T030000.MET.nc,GMAO_MERRA2.20160917T040000.MET.nc,anc_cor_file_28jan2014.nc,morel_fq.nc,aerosol_msis2a,gebco_ocssw_v2020.nc,gebco_ocssw_v2020.nc,gebco_ocssw_v2020.nc,mld_climatology_woa1994.hdf,20160917120000-CMC-L4_GHRSST-SSTfnd-CMC0.1deg-GLOB-v02.0-fv03.0.nc,20160917120000-CMC-L4_GHRSST-SSTfnd-CMC0.1deg-GLOB-v02.0-fv03.0.nc,sss_climatology_woa2009.hdf,no2_climatology_v2013.hdf,alpha510_climatology.hdf,taua865_climatology.hdf,calcite_table-20170109.txt,owmc_lut.hdf,water_spectra.nc" ;
:calibration_data = "" ;
:mask_names = "ATMFAIL,LAND,CLDICE,HILT" ;

:software_name = "l2gen" ;
:software_version = "9.7.0-V2024.5" ;
:input_sources = "manifest.safe,GMAO_MERRA2.20160917T020000.MET.nc,GMAO_MERRA2.20160917T030000.MET.nc,GMAO_MERRA2.20160917T040000.MET.nc,GMAO_MERRA2.20160917T020000.MET.nc,GMAO_MERRA2.20160917T030000.MET.nc,GMAO_MERRA2.20160917T040000.MET.nc,anc_cor_file_28jan2014.nc,morel_fq.nc,aerosol_msis2a,gebco_ocssw_v2020.nc,gebco_ocssw_v2020.nc,gebco_ocssw_v2020.nc,mld_climatology_woa1994.hdf,20160917120000-CMC-L4_GHRSST-SSTfnd-CMC0.1deg-GLOB-v02.0-fv03.0.nc,20160917120000-CMC-L4_GHRSST-SSTfnd-CMC0.1deg-GLOB-v02.0-fv03.0.nc,sss_climatology_woa2009.hdf,no2_climatology_v2013.hdf,alpha510_climatology.hdf,taua865_climatology.hdf,calcite_table-20170109.txt,owmc_lut.hdf,water_spectra.nc" ;
:calibration_data = "" ;
:mask_names = "ATMFAIL,LAND,CLDICE,HILT" ;

[wxfei19]

I am sorry, I find the attached file could not show in this post.
so I want to upload the jpg files

[OB SeaDAS - xuanyang02]

We did make changes to our MSI processors. V2025.5 should be better.