When we compute indices we should document which tag/release of the code was used for reproducibility/replicability. For example, if we make a processing run in March using the latest version of this code and then make another processing run in June then the assumption may be that the two versions of the code are identical when this isn't necessarily the case. There should be some mechanism put in place to allow us to get a version or tag number of the package being used and then write that into the result/output files containing the index datasets. This metadata can be added to the NetCDF files as global and/or variable level attributes. The attributes can contain the git commit information in order to most accurately identify the version of the package being used.