Changeset 1926 for trunk/LMDZ.TITAN/README

Timestamp:

May 28, 2018, 11:47:03 PM (7 years ago)

Author:

jvatant

Message:

1) Microphysics diags / outputs :

---

+ Add supplementary diagnostics outputs for microphysics ( precip, flux, rc ... ) ( new muphy_diag.F90 module )
+ Correct the outputs of microphys tracers to be in X/m-3 to be comparable to "standard values"

+ Also update the deftank callphys.def with latest revs modifs for microphysics

2) Condensation / chemistry updates :

---

+ Moved chemistry AFTER microphysics

• To have mufi condensation before photochem
• Chemistry called last coherent with the fact that it brings back fields to equilibrium

+ If 2D chemistry, make zonally averaged fields go through mufi and chem condensation

to have non saturated profiles in input of photochemistry
( other 'short' processes neglected as 2D -> no diurnal cycle, just seasonal evolution )

+ Also corrected the positivity check ( took Mars GCM syntax ) after chemistry ( could previously lead to negs )

3) Noticed a weird behaviour ( bug? ) :

---

+ If generalize the use of arrays *_indx for tracers, to get rid of ugly "iq+nmicro",

it ends up with weird results / crash in optim mode ( ok in debug ) but didn't find out why ...

--JVO

File:

• trunk/LMDZ.TITAN/README (modified) (1 diff)

trunk/LMDZ.TITAN/README

@@ -1438 +1438 @@
 + Also modifs to have chemistry in 1D in rcm1d ( and moved gr_kim_vervack in phytitan to be accessible for 1d )
 + In chemistry added a check.c to verify coherence of sizes between C and Fortran
+
+== 28/05/18 == JVO
+1) Microphysics diags / outputs :
+--------------------------------
++ Add supplementary diagnostics outputs for microphysics ( precip, flux, rc ... ) ( new muphy_diag.F90 module )
++ Correct the outputs of microphys tracers to be in X/m-3 to be comparable to "standard values"
+
++ Also update the deftank callphys.def with past revs modifs of microphysics
+
+2) Condensation / chemistry updates :
+-----------------------------------
++ Moved chemistry AFTER microphysics 
+  - To have mufi condensation before photochem
+  - Chemistry called last coherent with the fact that it brings back fields to equilibrium
++ If 2D chemistry, make zonally averaged fields go through mufi and chem condensation
+  to have non saturated profiles in input of photochemistry
+  ( other 'short' processes neglected as 2D -> no diurnal cycle, just seasonal evolution )
++ Also corrected the positivity check ( took Mars GCM syntax ) after chemistry ( could previously lead to negs ) 
+
+3) Noticed a weird behaviour ( bug? ) :
+---------------------------------------
++ If generalize the use of arrays ***_indx for tracers, to get rid of ugly "iq+nmicro",
+  it ends up with weird results / crash in optim mode ( ok in debug ) but didn't find out why ...