GitLab

ICOADS quality control indicator

• The irf variable stands for intermediate reject flag. A value of 2 indicates that the data source or report lacks suitable quality. This data is rejected from the ship data selection in split_by_type.R

• A deck priority is assigned in add_dck_priority.R .

  A certain priority is assign to some decks which are known to have a good quality data. The data expected to be of best quality is assigned a priority of 1, data with larger priority numbers will be flagged as the worst duplicate if reports are identified as potential matches. These priority values are based on those from ICOADS.

Met Office climatological check

A precision criteria on selected variables is used to perform a climatological quality control on the data.

The precision criteria for each key variable per dck, year and/or sid can be found in Table 5 of the technical report.

Having a precision criteria not only preforms quality control on the data but helps to set tolerances when comparing variables from suspected duplicates.

A comparison of climatological variables allows for a match between reports in the duplicate identification procedure done by get_pairs.R.

Met Office track check

Several dck have id's that indicate a logbook, sheet or other block of data that can be linked together to form ship tracks.

The linked id's are checked using the Met Office Quality Control track check (MOQC track check from IMMA) as well as for time duplicates.

Reports that fail the track check are flag as the worst duplicate. Where positions (lat, lon) are similar the best duplicate is select by dck priority and number of quality variables found by the climatological check. The track check is also perform in the same get_pairs.R script.