CDO

Hi,

I am post-treating the reanalysis of NOAA_20CR (ftp://ftp.cdc.noaa.gov/Datasets/20thC_ReanV2/Dailies) from 1871 to 2010. I have a problem with the variable hgt (geopotential height).
I get the files from the site from year_ini to year_end. Then I create a file using cdo mergetime and then I apply different command to uncompress data (-b f32), to change the initial name of the variable, to extract the level 500hPa.
I have a problem when I use -b F32 and I do not know why (see the attached script process_NOAA_20CR_daily.sh_example) because I did not have any problems with the other variables I downloaded:
cdo -b F32 chname,hgt,zg file_temp_daily1.nc ${file_fin_daily_3D}
I saw on internet that the error is linked to -b F32 or -b F64. If I put -b F64 it seems to work. But I do not want to have my data in double as all the other reanalysis I already post-treated are in float.
At the moment I am creating the global file from 1871 to 2010 without uncompressing the data, and I will use an NCL program to write a final file with data in float but it is not very convenient for me.

I attached the script process_NOAA_20CR_daily.sh_example and the script to get the initial 3D files on the ftp site get_daily_NOAA_20CR_3D_pressure.ksh. I put arbitrary BEGIN_YEAR=1871 and END_YEAR=1875 in the script process_NOAA_20CR_daily.sh_example.

Do you understand why it is not working ?

Thanks,
Regards,
Laure