Next, we'll run GSID2MPR on the command line using the following command:
input/gsi_data/diag_conv_ges.mem* \
-outdir $MET_TUTORIAL_DATA/output/gsid2mpr \
-swap
Here, we've reformatted five conventional GSI diagnostic files. The -swap option indicates that the Endianness of the data must be switched. For each input file, a .stat output file is written to the directory specified by the -outdir option. Notice in the log messages that it checks for and skips duplicate entries in the data. The -no_check_dup option disables the checking for duplicate data. Next, run a similar command but for satellite radiance data:
input/gsi_data/diag_amsua_n18_ges.mem* \
-outdir $MET_TUTORIAL_DATA/output/gsid2mpr \
-swap
Next, run very similar commands using the GSIDENS2ORANK tool to process the conventional and satellite radiance data as a 5-member ensemble, rather than as indivual files. This generates ORANK output lines instead of MPR lines:
input/gsi_data/diag_conv_ges.mem* \
-out $MET_TUTORIAL_DATA/output/gsidens2orank/diag_conv_ORANK.stat \
-swap
input/gsi_data/diag_amsua_n18_ges.mem* \
-out $MET_TUTORIAL_DATA/output/gsidens2orank/diag_amsua_ORANK.stat \