Adding cloud-related data to Vtable.GFS

[noteinthepad]

I am attempting to run WPS for a later WRF Solar simulation using GFS data as an input, however, after looking at the Vtable.GFS, I couldn't find any entries listed for cloud-related variables (cloud fraction, cloud water mixing etc).

I assume I have to edit the Vtable to fit my needs, but reading through the user's guide, it says the metgrid Name column must match with entries in METGRID.TBL, but reading through the file, there aren't any matching variables (TCDC, CWAT etc)?

Am I missing something here? WPS cannot take such data as an input?

 

[kwerner]

Hi,
I believe all cloud-related variables are initialized during the real.exe process, and then calculated later by the physics schemes in wrf. So it may not be necessary to modify anything during WPS.

 

[noteinthepad]

Hello,
Thank you for your reply.

I found that cloud-related variables are available in the wrfout files. However, without initial cloud data inputs, wouldn’t the simulated clouds lack accuracy due to the absence of initial cloud cover data?

Also, is there documentation explaining the variable names in METGRID.TBL? The shorthand names (e.g., CLDMASK, QNOCBB2D) are hard to interpret.

I’d greatly appreciate your support.

 

[meteoadriatic]

There is no much sense to use simulated clouds from input data to initialize cloud field in the WRF. WRF parameterizes clouds immediatelly from the relative humidity and other data and starting point is not that much important. With that being said, there are two things that should be noted:

1) There is insert_init_cloud namelist.input entry, for the real.exe to generate clouds before wrf.exe starts. Wheather or not this gives better simulation is not clear. See here Insert initial clouds by gthompsnWRF · Pull Request #1323 · wrf-model/WRF

2) There are techniques to asimilate clouds, not model cloud fields but real cloulds (satellite measurements). See for example https://www.arl.noaa.gov/documents/...heme2_Biazar_Cloud_attenuation_photolysis.pdf

 

[kwerner]

without initial cloud data inputs, wouldn’t the simulated clouds lack accuracy due to the absence of initial cloud cover data?

I believe the model uses other known variables to calculate cloud cover at the initial time.

Also, is there documentation explaining the variable names in METGRID.TBL? The shorthand names (e.g., CLDMASK, QNOCBB2D) are hard to interpret.

That's a good question. I've reached out to the developer of the metgrid program to see if they can point us to any existing documentation. I'll get back to you when I get a response from them.

 

[haiqingsong]

There is no much sense to use simulated clouds from input data to initialize cloud field in the WRF. WRF parameterizes clouds immediatelly from the relative humidity and other data and starting point is not that much important. With that being said, there are two things that should be noted:

1) There is insert_init_cloud namelist.input entry, for the real.exe to generate clouds before wrf.exe starts. Wheather or not this gives better simulation is not clear. See here Insert initial clouds by gthompsnWRF · Pull Request #1323 · wrf-model/WRF

2) There are techniques to asimilate clouds, not model cloud fields but real cloulds (satellite measurements). See for example https://www.arl.noaa.gov/documents/...heme2_Biazar_Cloud_attenuation_photolysis.pdf

GFS has QC.

 

[kwerner]

@meteoadriatic
To your question regarding METGRID.TBL documentation, the metgrid program doesn't know anything about fields -- it simply horizontally interpolates what it's given in intermediate files using the methods described by entries in the METGRID.TBL file. However, if the metgrid program produces an interpolated field (e.g., CLDMASK), then the real.exe program would generally need to know about that field, and so those fields (CLDMASK, QNOCBB2D) should have entries in WRF's registry file, ideally with descriptions. For example:

state real  qnocbb2d   ij   misc    1    -   i01{17}rhdu "QNOCBB2D"   "Surface organic carbon biomass burning number conc emission"  "kg-1 s-1"