Does the output variable W in the wrfout file include the vertical motion estimated by the convective parameterization?

[Xipeng Jin]

Does the output variable W (described as the z-wind component) in the wrfout file include the parameterized vertical motion (eg. convection parameterization, when it is on) ? Or is it just the vertical velocity of the model's resolved scale? What is the meaning of the variables "w, ww, rw, ww1, ww_m, wwp, rw_tend, rw_tendf, w_save, w_subs, w_save" in the Registry.EM_COMMON？
How is the vertical velocity w calculated? Which .F files are the relevant codes in?

 

[Ming Chen]

Xipeng,
You can find their definitions in Registry.EEM_COMMON, for example,
"w" in wrfout is vertical velocity on the resolved scale of WRF.
"ww" is vertical velocity that is mu-coupled on model level (deta_level/dz)
"rw' is mu-coupled w ( i.e., (mu+mub)*w )
...
I have answered your question of w calculation in your other post.

 

[Xipeng Jin]

Ming,
Thanks a lot! I still have three questions：
1) According to kwerner's reply to (How is vertical velocity defined?): 'w' is the true vertical velocity by its conventional definition (dz/dt). WRF has another variable (ww) that is used for advection and which accounts for the coordinate, w seems to contain both advective and parametric vertical motion (when there is a parametric scheme turned on). Can it be understood this way?
2) How is the effect of the parameterization scheme passed/updated to atmospheric variables such as u, v, t, q? Is the update of w different from these variables, is it mainly obtained by indirect diagnosis?
3) What does " resolved scale of WRF" mean? Does it only refer to grid scale (dx, dy, dz) or also include parametric parts?

 

[Ming Chen]

Xipeng,
Please see my answers below:

1) According to kwerner's reply to (How is vertical velocity defined?): 'w' is the true vertical velocity by its conventional definition (dz/dt). WRF has another variable (ww) that is used for advection and which accounts for the coordinate, w seems to contain both advective and parametric vertical motion (when there is a parametric scheme turned on). Can it be understood this way?

That is right.

2) How is the effect of the parameterization scheme passed/updated to atmospheric variables such as u, v, t, q? Is the update of w different from these variables, is it mainly obtained by indirect diagnosis?

The codes related to update of variables are:
dyn_em/module_big_step_utilities_em.F
dyn_em/module_em.F
dyn_em/solve_em.F

Specifically, 'rk_tendency' in dyn_em/module_em.F handles the tendency update.

'w' is a diagnostic variable.

3) What does " resolved scale of WRF" mean? Does it only refer to grid scale (dx, dy, dz) or also include parametric parts?

Resolved-scale refers to the grid-scale of WRF. Some sub-grid scale processes such as cumulus must be parameterized.
Regarding 'w' calculation, the impacts of parameterization is implicitly included.

 

[Xipeng Jin]

Ming,

Thank you very much. Your reply has completely solved my confusion！

Xipeng

 

[Kyohei]

Hi Ming or the staff member,

I have further question regarding the posts here.
From the above posts, I understand that ww [Pa/s] is used only for the advection (=does not include parametric vertical motion), and w[m/s] contains both advective and parametric vertical motion.
However, I want to find out the vertical velocity** in [m/s] which only contains advective vertical motion. Currently, I'm trying to find some way to convert ww[Pa/s] into the vertical velocity** which only includes advection, but I can not understand all the coordinate transformation stuff from the eta coordinate to conventional coordinate.
(the vertical velocity**) I'm putting " ** " just to differentiate from the conventional vertical velocity.

Thus, the question is;
Q1. Is there any equation to convert ww[Pa/s] into the vertical verlocity**[m/s] which only includes advection?

Furthermore, from the previous post below by Ming, I have another question.

3) What does " resolved scale of WRF" mean? Does it only refer to grid scale (dx, dy, dz) or also include parametric parts?

Resolved-scale refers to the grid-scale of WRF. Some sub-grid scale processes such as cumulus must be parameterized.
Regarding 'w' calculation, the impacts of parameterization is implicitly included.

Q2. (I am currently using Building Energy Model and Building Effect parameterization, and) are these also included in the "parametric parts" of the conventional vertical velocity 'w'?