Better SWE validation dataset.

[dlawrenncar]

Andy Wood in TSS at NCAR suggests:

Hi all,
Related to the ILAMB SWE, it currently appears to use a daily 1 degree product, CanSISE (https://nsidc.org/data/nsidc-0668/versions/2), that merges 5 products including GLDAS, ERA5Interim, MERRA, Crocus and GlobSnow. There is a new paper out (a global snow intercomparison from SnowPEX) that supports my thought that we can find better SWE datasets for ILAMB.

https://tc.copernicus.org/articles/19/201/2025/

Among these 5, GLDAS is problematic and ERA5Interim is also not strong (we knew this), while the others are slightly better with Crocus performing the best of the 5. But ERA5Land, which we've been using for other modeling work at NCAR, looks like an even better option, and at 0.1d/hourly it will enable the smaller scale analyses that are useful for land/hydrology model diagnostics. I would suggest picking several of the better performing analyses as new ground truths to make available in ILAMB. I have CONUS-wide SWE from era5land in our campaign storage, as well as download routines so that we can pull the global set if needed.

Cheers,
Andy