Hi Gabe, Indeed you did the proper modification, but this requires a few additional changes in level1_landsat.py and params.py. See the following diff (I will include these changes in further releases of Polymer) : diff --git a/polymer/level1_landsat8.py b/polymer/level1_landsat8.py index a70ad72..b...

Dear Binbin,

You need to provide the granule directory like "S2A_OPER_MSI_L1C_TL_SGS__20160912T204513_A006400_T19TDJ_N02.04" , not IMG_DATA.
Please see section 2.3.4 of the README.
Cheers,
François

Dear Binbin, You need to provide credentials to NASA EarthData to access ancillary data. As mentioned in the error message: Error authenticating to NASA EarthData for downloading ancillary data. Please provide authentication through .netrc. See more information on https://support.earthdata.nasa.gov/...

Hi, It seems that the download of "auxdata/common/no2_climatology.hdf" has failed, could you try to remove this file and try again ? By the way, there was a letter "i" missing in the filename you mentioned. Concerning the processing of Landsat8 files : * You need to provide a directory name, not a t...

Dear Vishnu, Bitmask is coded in binary, where each bit represents a flag, as coded in bitmask's attributes, from LAND=1=2^0 to INCONSISTENCY=2048=2^11. Thus 2112 = 64+2048 = THICK_AEROSOL+INCONSISTENCY It is recommended to remove pixels where any of the first eleven bits is raised, thus where "bitm...

So, for the record, the patch is: diff --git a/polymer/level1_msi.py b/polymer/level1_msi.py index e2a7b63..c3e1926 100644 --- a/polymer/level1_msi.py +++ b/polymer/level1_msi.py @@ -198,8 +198,8 @@ class Level1_MSI(Level1_base): XDIM = int(e.find('XDIM').text) YDIM = int(e.find('YDIM').text) - X, Y...

Great! Good to read that it solves the issue. I will integrate this modification in further Polymer releases. Just a minor comment: in python3 you would have to use XDIM//2 to stay with integers, XDIM/2 gives a float. But I don't expect that it should affect the results - maybe the performance ? But...

Hi Martin, Thanks for the detailed feedback ! You are right, it is clear Polymer is wrong in how it reads the lats/lons. As you say, the coordinates of the centre of the north east pixel should not be the same at different resolutions. Looking at the code (method init_latlon in level1_msi.py), it is...

I doubt your command has succeeded, because the command should be "make", in the directory containing the makefile, not "make ./polymer4.13/".

The makefile is used by the command "make".