2024-02-16 - February 16, 2024

[dprincz]

Updates

• @AlaBahrami ran the grid-based version using r1860-ME
• Mohamed added Zelalem's code to his private repository, which confirms there are additional changes related to FROZENSOILINFILFLAG and MOUNTAINMESH from r1860-ME
• @dprincz continues to integrate r1860-ME into MESH-Dev; added @mee067 as a 'collaborator' and tracking questions about migrating specific features using 'issues'
• @AlaBahrami and @kasra-keshavarz opened an issue for the feature request for reading successive series of driving data sets

Discussion

• Demonstrated GitHub wiki pages; opportunity to cull out irrelevant materials and/or conflicting information
• @mee067 notes might want to keep migrated pages on the UofS wiki for old references, updated with a link to the GitHub wiki with updated information
• @dprincz added @mee067 and @fuadyassin to fork of MESH-Dev to track integration of r1860-ME, using issues to communicate where questions about specific changes in the code; at changes for resume files, changes for NetCDF vector files, 10m snow limit in CLASS, routing updates, then format updates for output files remain
• Geofabric: St-Mary/Milk has 18 separated basins (e.g., basins where flow doesn't cross between boundaries)
• Workflow had indexing issue which has since been resolved; an option to add an additional field to the db caused reindexing of the basins that derived invalid RANK/NEXT connections creating invalid derived drainage directions
• Non-contributing area: They still exist, their outlet is just disconnected from the main outlet; can be masked out using SUBBASINFLAG if computationally exhaustive (doesn't seem so), these areas do not contain blue lines/routing properties (using Transboundary Geospatial Fabric, TGL).. Would have to be modified for them to contribute vertical balance components (PREC/ET/ROF) to an adjacent subbasin as required
• MERIT-Hydro likely has the same for non-contributing areas (e.g., no blue lines), but also has costal hillslope subbasins that do have river properties but do not flow into any other basins (e.g., to ocean)
• Note: MERIT-Hydro does not contain lakes
• @dprincz: In technicality this isn't an issue; having the 'island' basins permits future work if contributions from those elements should be explored
• Regarding inconsistent dimension names between versions, the workflow can be configured to switch based on some versions; in the code, the aim is current code should be able to accommodate all former options (this is in pending changes to compare dimension/variable names to lists, as opposed to a single variable name)

Suggestions

• Combining/aggregating/merging too small subbasins that would invalidate physics (e.g., blowing snow, where the stretch of a subbasin is smaller than the fetch difference) -- @kasra-keshavarz This is in-progress in the model-agnostic workflow
• Using dataset to check if those areas are lakes before aggregation
• Configuration options via YAML? @AlaBahrami will make a new 'issue' for the suggestion

Standing items

• Al will keep in touch with Mohamed Ahmed regarding variable-contributing area code

Actions

• @dprincz continuing migration of r1860-ME into MESH-Dev with @mee067