Checks Overview
ESGF-QC is built from many atomic checks, grouped by category. The plugin just decides which ones to run; each category focuses on a specific aspect of file quality.
Dimension Checks
Verify that required dimensions exist, have a positive/expected size, and match the declared shapes of variables.
Variable Checks
Ensure variables are present, their shape matches their dimensions, and (when bounds are declared) each value lies within its bounds.
Time Checks
Focus on the time axis: presence of bounds with the right shape, and consistency between the time axis and the filename’s time range.
Attribute Checks
Check that mandatory global and variable attributes are there, correctly typed/encoded, and their values are valid against regex patterns or controlled vocabularies.
DRS & Consistency Checks
Validate filenames and directory paths and compare with the netcdf file's attributes; verify consistency between attributes like frequency vs table_id, experiment_id, institution, variant_label, etc., against project rules/CVs.
Coming Next
- Data plausibility checks (e.g. physical plausible, statisctically meaningful, ...).
- Dataset-level checks across multiple files (time continuity, no gaps/overlaps, consistent metadata ).
Detailed Inventory
For the full list of checks (IDs, severities), see the spreadsheet :