UQ calcs for CARRA Scripts used to do the odb data extraction and data processing for the uncertainty estimation of CARRA
-
get_all_members: copy the odb_stuff.tar files for all members in relevant dates. This can take several days depending on the load. Will fetch the tar balls and untar then. It also calls dcagen.
-
get_logfiles.sh: copy the logfiles.tar files containing the "HM_Date*" files
-
callodbsql_allmems.sh: call the odbsql commands for each member, using search_HM_Date.py for each date. The result from odbsql is stored under each YYYY/MM/DD/init/mbrXXX with name mbrXXX_Codetype_Obstype_Varno.dat
-
search_HM_Date.py reads the HM_Date log files, identifies the relevant pairs of Obstype/Codetype/Varno and calls the odbsql commands for each combination. A summary of the variables found is stored under YYYY/MM/DD/out_hm_search.txt. There is a also a list of the odbsql commands to be used.
-
call_procodb.sh: call the script process_odb.py for all dates and Obstype/Codetype/Varno combinarion. It scans the relevant file for each member and prints a summary for each init time under YYYY/MM/DD/summary_Codetype_Obstype_Varno.txt