Creating CLM surface datasets
Comments
@Ruby Leung, I will ask around a bit, but I don't think the land and ocean mask can be consistent. Meaning that there will always be some points on the surface of the earth that will be covered by both land and ocean and some points that might have neither! The land is using quad, the ocean is using hexagons – there will always be some misalignment. Even with using POP, the mesh in the Northern Hemisphere was a twisted lat/lon mesh so it could no "line up" with regular lat/lon meshes. That said, we can all still use the same observational dataset to produce our elevations/bathymetry. We tend use ETOPO2. Cheers, Todd.
@Todd Ringler (Unlicensed), @Ruby Leung: Apart from supporting a regular lat/lon grid, the land model is now capable of running on unstructured grid. For the unstructured grid, the land model does not impose any restriction on the shape of grids. Thus, one could even run the land model on a MPAS grid.
Subject: Land-sea mask
We are in the process of generating subgrid topographic landunits for the atmosphere and land model. The fallback plan is to run both atmosphere and land on the 1-degree CAM-SE grid with subgrid topography defined using a local approach, with up to 12 classes per grid. To ensure that the land-sea mask is identical for the atmosphere, land, and ocean models, we need to coordinate the development of the land-sea mask dataset.
For the atmosphere and land models, the subgrid topography information is being derived from a global DEM data at 90-m resolution, so the land-sea mask will be determined from this dataset.
What is the plan for the ocean model? Has a land-sea mask been created yet? Should we coordinate with @Todd Ringler (Unlicensed)?
Ruby