Profiling month 1 vs month 2

I've copied u-ae771 to u-af925 and added in Dr Hook for this profiling.

Profiling the top routines

1st month

Routines
UM_SHELL (5,090s)
U_MODEL_4A (5,086s)
ATM_STEP_4A* (4,750s)
ATMOS _PHYS- ICS1 (906s) ATMOS _PHYS- ICS2 (270s) EG_ SL_ HELM- HOLTZ (227s) TR_ SET_ PHYS _4A* (229s) EG_CORRECT _TRACERS _PRIESTLEY (293s) SL_ TRAC- ER1_ 4A (391s) EG_ SL_ MOI- STURE (54s) EG_SL_ FULL_WIND (120s)  ⇓  UKCA_MAIN1 (992s)
Profile for ATMOS_ PHYSICS1 Profile for ATMOS_ PHYSICS2 and EG_SL_ HELMHOLTZ EG_SL_WIND_U, EG_SL_WIND_V & EG_SL_WIND_W (14 + 13 + 47 = 74s) STASH (280s) See profiling for UKCA_ MAIN1
Itself (78s) EG_INTERPOLATION _ETA (253s) DEP- ARTURE_ POINT _ETA (100s) STWORK (279s)
EG_ CUBIC_ LAG- RANGE (78s, itself) MONO_ ENFORCE (29s, itself) Itself (119s) Itself (8s) PP_ HEAD (128s) EXP- PXI (101s, itself)
*should also link to SWAP_BOUNDS, like many other returns.

2nd month

Routines
UM_SHELL (5,604s)
U_MODEL_4A (5,590s)
ATM_STEP_4A* (5,197s)
ATMOS _PHYS- ICS1 (897s) ATMOS _PHYS- ICS2 (266s) EG_ SL_ HELM- HOLTZ (234s) TR_ SET_ PHYS _4A* (214s) EG_CORRECT _TRACERS _PRIESTLEY (292s) SL_ TRAC- ER1_ 4A (383s) EG_ SL_ MOI- STURE (55s) EG_SL_ FULL_WIND (116s)  ⇓  UKCA_MAIN1 (991s)
Profile for ATMOS_ PHYSICS1 Profile for ATMOS_ PHYSICS2 and EG_SL_ HELMHOLTZ EG_SL_WIND_U, EG_SL_WIND_V & EG_SL_WIND_W (14 + 13 + 37 = 64s) STASH (291s) See profiling for UKCA_ MAIN1
Itself (89s) EG_INTERPOLATION _ETA (249s) DEP- ARTURE_ POINT _ETA (88s) STWORK (291s)
EG_ CUBIC_ LAG- RANGE (78s, itself) MONO_ ENFORCE (29s, itself) Itself (116s) Itself (8s) PP_ HEAD (128s) EXP- PXI (101s, itself)
*should also link to SWAP_BOUNDS, like many other returns.

Summary

Not a lot of difference between months, and similar to my profile for the first month of prototpye UKESM0.4.

Profiling UKCA_MAIN1

2nd month

Routines
UKCA_MAIN* (991s)
UKCA_AERO_CTL (147s) UKCA_ ACT- IVATE (40s) UKCA_CHEMISTRY_CTL (294s) UKCA_FASTJX (153s) UKCA_ NEW_ EMISS_ CTL (77s)
UKCA_AERO_STEP (142s) UKCA_ ABDUL- RAZZAK _GHAN (38s) ASAD_CDRIVE (261s) FASTJX_PHOTOJ (152s)
UKCA_ COAGWITH- NUCL (40s) UKCA_ CONDEN (39s) Itself (37s) ASAD_SPMJPDRIV (232s) FASTJX_OPMIE (92s) FLINT (28s, itself) Itself (22s)
Itself (31s) UKCA_ COND_ COFF _V (31s, itself) ASAD_SPIMPMJP (228s) FASTJX_ MIESCT (66s) Itself (26s)
SPLINSLV2 (107s, itself) SPFULJAC (67s, itself) Itself (21s) BLKSLV (66s)
Itself (45s)

These are similar timings to the first month of our prototype UKESM0.4, so it's clear that the UKCA_CHEMISTRY_CTL section is taking much longer than normal for our prototpye UKESM0.4.