This post was from a previous version of the WRF&MPAS-A Support Forum. New replies have been disabled and if you have follow up questions related to this post, then please start a new thread from the forum home page.
mgduda said:When you do have a chance to try the RAP data again, I'd be interested to see exactly what the log file looks like. Regardless, I think the process of creating ICs and LBCs from RAP data will fail because the MPAS-Atmosphere v7.0 code doesn't have support for input datasets given on a Lambert conformal map projection. I have some changes in a branch named 'init_atmosphere/lambert_conformal_support' of my fork of the MPAS-Model repository that should add support for this projection; specifically, the changes are in this commit. Depending on your level of familiarity with git, I can provide more detailed instructions on how to merge these changes into your working copy of the MPAS-Model code.
git remote add mgduda https://github.com/mgduda/MPAS-Model.git
git fetch mgduda
git merge mgduda/init_atmosphere/lambert_conformal_supportUpdating 51d5624..95e7e46
Fast-forward
src/core_init_atmosphere/mpas_init_atm_cases.F | 72 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++----
src/core_init_atmosphere/mpas_init_atm_surface.F | 17 +++++++++++++++++
2 files changed, 85 insertions(+), 4 deletions(-)mgduda said:Assuming you've obtained the MPAS-Model code by cloning the repository, I think the following should be sufficient:
If the merge of the Lambert conformal changes is successful, you should see something like the following:Code:git remote add mgduda https://github.com/mgduda/MPAS-Model.git git fetch mgduda git merge mgduda/init_atmosphere/lambert_conformal_support
Code:Updating 51d5624..95e7e46 Fast-forward src/core_init_atmosphere/mpas_init_atm_cases.F | 72 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++---- src/core_init_atmosphere/mpas_init_atm_surface.F | 17 +++++++++++++++++ 2 files changed, 85 insertions(+), 4 deletions(-)
mgduda said:The message "Error in interpolation of st_fg to MPAS grid: num_st = XXXX" is generally indicative of an issue with the soil temperature. Would it be possible to use some of the utilities that come with the WRF Pre-processing System (WPS) to check that there are valid soil temperature data in your intermediate file? Specifically, you could use the "rd_intermediate" utility to check that there are fields named, e.g., ST000010, and you could use the "int2nc" utility to convert the intermediate file to netCDF format, where it would be easier to view the soil fields with, e.g., ncview.
Alternatively, if you could attach your complete log.init_atmosphere.* files, I can take a look to see what soil data appear to have been read and interpolated.
mgduda said:Thanks very much for checking. I think this is the problem -- that we have hard-wired logic for known soil layers to set, e.g., the soil layer bottom depth based on the name of the field (e.g., ST000010), and we don't currently have code to handle RAP soil layers.
It shouldn't be terribly difficult to add additional logic to handle RAP soil layers, though I couldn't say when I might be able to implement this myself. If you'd like to have a go at implementing support for RAP soil data yourself, I could probably provide some guidance. The best approach may be to open a separate thread in the "Code development" section of the forum.
As another option, it should in principle be possible to use land-surface fields, e.g., from the GFS model with RAP atmospheric fields. There could be issues of consistency in doing this, but it should be possible in any case.
mgduda said:If the test simulation looks good, and if you'd be interested in getting your changes back into the main MPAS-Model repository, just let me know and I can help in the process of creating a new pull request from your changes.
mgduda said:If the test simulation looks good, and if you'd be interested in getting your changes back into the main MPAS-Model repository, just let me know and I can help in the process of creating a new pull request from your changes.