Simulations over land/topography

[sjboeing]

No description provided.

[sjboeing]

• Use of local vs. averaged profiles for e.g. nudging
• Dealing with points below/near topography in gradient calculations
• Input parameters for land-surface model

[sjboeing]

Will be adding some example files for Cardington soon

[sjboeing]

Need to think about what height is 0 for MONC inputs.

[sjboeing]

@leifdenby: I think this might be best placed as part of the conversion process. For MONC, the local surface is at 0.

[sjboeing]

@leifdenby: I now started working on simulations over land again. One of the things I am planning to change is the sampling_mask in the forcings input definition. I would like to give users the choice between "all", "ocean_only" and "land_only". I am not sure if there should be a default choice (all?), or if we should force users to specify it. The latter may be the better option, as it prevents users being unaware of the options. What do you think?
While looking into this, I came across the following in the forcing INPUT_REQUIRED_FIELDS: time_sampling_method=(None, "str"). I suppose this is correct, and should not be time_sampling_method=(None, str), but let me know if this needs to change.

[sjboeing]

Many required changes are in #125. What needs to be added still is a way to ensure the surface is placed at the right level. Thinking about this a bit more today, I think this should already be done before conversion, as we will get the maximum amount of near-surface information by choosing the right height levels at this point.

[leifdenby]

@sjboeing can we close this issue? Or are there still issues to resolved with regarding creating forcings for simulation over land?

[sjboeing]

@leifdenby: I think we should not do anything before 0.1.0 anymore, but may want to do a fix for non-zero surface height in 0.2.0