Where the conventions come from

Three kinds of fact go into a run, and they come from three places.

The ISMIP7 data request, from isschecker

The state and flux scalars this package writes (lim, limnsw, iareagr, iareafl and the six tend fluxes) are ISMIP7 variables in their own right, and the data request fixes their standard names, units and long names. That request is maintained in ISM_SimulationChecker and shipped in its isschecker package.

This package reads the request from isschecker at runtime rather than keeping a copy, which is why isschecker is a dependency even though none of its code runs. A copy once drifted, and the files this tool wrote were rejected by the compliance checker.

Two things follow:

• A correction to a variable’s name or units is released by isschecker, not here. Update isschecker and the files pick it up.

• A change to the shape of the request, such as a renamed column, needs a release here as well.

The sea-level scalars (slvaf, slg20, sla20) are derived here rather than submitted by a model, so they are not in the data request and carry no standard name.

The CORE experiment table, from this package

ISMIP7_experiments_CORE.csv, inside the package, records which CORE configid runs which scenario under which forcing ESM. The ensemble driver uses it to pair each projection with its historical run: C001 for CESM, C002 for MRI. Override it with ismip7-scalars-ensemble --core-csv.

The area factors and masks, from Globus

These are large gridded fields, distributed on the ISMIP Globus server under Output-Processing rather than packaged with anything. Point --datapath at them. File and directory conventions lists the names.