CCAPS (all flavours) - check all factors of rho in code and wiki

[JOThurgood]

No description provided.

[JOThurgood]

Found and corrected typos in the wiki.

[JOThurgood]

Currently in the variable density versions of the code the following happens in terms of factors of rho

The old ( "t=n" ) density is used in

• step 2: this is uncontroversial
• step 3: There is a question over whether advect_dens should occur before this (immediately after step_2) and then a half time density can be computed for use in this step (you would also need to flesh out the code for step 3 which assumes a lot of things don't change but now do with the different factor of rho). This is what seems to happen in lm_atm in pyro2.

The half time (t=n+1/2) density is used in:

• step 4: gravitational source /forcing for the beta/rho or 1/rho coefficient (seems appropriate). (NB only in the atmospheric code, the other versions have not been updated yet and use the n+1 value for rho).

The new (t=n+1) density is used in:

• step 4: for the 1/rho coefficient in front of the lagged pressure (this seems inappropriate)
• all of step 5: seems appropriate.

In other words:

• Should advect_dens be called before step_3, and used in the updated interface states ? (as it is in pyro2 lm_atm? Remember, step_3 would have to be fleshed out in this case - it basically assumes nothing changes except for a new up-winding.

• Should you use a half time rho for all coefficients in step_4? (i.e. not just for gravity in the case of the atmospheric version, and if so update other versions consistently.)