Should there be a standardized calibrated data file framework?

[FrankErrickson]

We currently calibrate the model to specific data sets with specific file formats (e.g. ocean heat content, global surface temperature anomalies, etc.). We could create a standardized file format for each calibration data set so users could swap in different calibration observations to run different sensitivity analyses.

However, this would only work for the model outputs currently being calibrated against (i.e. users could not add in a new dataset for model output that is currently not part of the model calibration without changing the source code).