grid%dz8w or grid%alt in solve_em.F is zero

[MSdueon12]

Hello moderator,
Please forgive me if the question is silly. I am struggling with solve_em.F, I knew this file aim to calculation the varaible in each time step. I just imitate the original solve_em.F to use other variables such as grid%ph_2,grid%t_phy and grid%dt. I want to use the variable dz8w for the distance between two vertical levels. However I found the value is zero, so is grid%alt. I use (grid%ph_2+grid%phb)/g, then the value is non-zero. I am quite confused. I attached the solve_em.F file and the part I am coding is just following CALL section.

I appreciate you help me out.

 

[Ming Chen]

Hi,
dz8w is not a state variable, which means it is not accessible in solve_em.F. This is why you see its value is zero.
However, grid%alt should not be zero in solve_em.F.
This question is related to WRF registry. I guess you need to understand how Registry works before you can modify WRF codes. Please take a look at the document here: https://www2.mmm.ucar.edu/wrf/WG2/software_2.0/registry_schaffer.pdf
Hope this is helpful for you.

 

[MSdueon12]

Hi,
dz8w is not a state variable, which means it is not accessible in solve_em.F. This is why you see its value is zero.
However, grid%alt should not be zero in solve_em.F.
This question is related to WRF registry. I guess you need to understand how Registry works before you can modify WRF codes. Please take a look at the document here: https://www2.mmm.ucar.edu/wrf/WG2/software_2.0/registry_schaffer.pdf
Hope this is helpful for you.

Thank you for your document, May I ask if I can use (grid%phb(i,k+1,j)+grid%ph_2(i,k+1,j))/g - (grid%phb(i,k,j)grid%ph_2(i,k,j))/g to represent dz8w in solve_em.F?

 

[MSdueon12]

From the document https://www2.mmm.ucar.edu/wrf/WG2/software_2.0/registry_schaffer.pdf, I am really confused that if all the variables who has the
Entries of the registry State can be used in Solve_em.F, with the formation of grid%name, How to distinguish which can be used, I used (grid%phb(i,k+1,j)+grid%ph_2(i,k+1,j))/g - (grid%phb(i,k,j)grid%ph_2(i,k,j))/g to get the distance between two layers but I found the perturbation geopotential is ph in Registry.EM_COMMOM, I don't know the if how can I use it.