Opened 5 years ago

Closed 5 years ago

#1194 closed help (invalid)

UKCA tutorial task5

Reported by: s1251469 Owned by: luke
Component: UM Model Keywords: ukca tutorial
Cc: Platform: HECToR
UM Version: 8.2


Thank you for the help on task 5.2 already. I have now managed to successfully get emissions into the ALICE. But i now also seem to have non-zero values in BOB.

When I first realised this, I found a bug in my ukca_chem_strattop.F90 file on lines 144,145. I had ALICE in slot 77 and BOB in slot 76. I corrected that and resubmitted but it didn't seem to correct anything.

Do you have any ideas what silly mistake I may have made?

many thanks

Change History (8)

comment:1 Changed 5 years ago by s1251469

PS. The job I'm using is xjecb

comment:2 Changed 5 years ago by luke

  • Owner changed from um_support to luke
  • Status changed from new to accepted

You do have values in 34065, but these are at the 10^(-163) level, i.e. noise. Tracers would appear at the 10^(-6) to 10^(-15) level. Looking in your 34064 tracer (ALICE) the smallest number isn't zero but at 10^(-185).

I'll take a copy of your job and have a look.


Last edited 5 years ago by luke (previous) (diff)

comment:3 Changed 5 years ago by s1251469

O yes. I didn't look at the magnitudes of BOB. Is it possible that my original error in ukca_chem_strattrop.F90 never got properly reinitialised and has left remnants?

It sounds like it should be ok just to carry on with the tutorial?

many thanks,

comment:4 Changed 5 years ago by luke

You are initialising the tracers to zero (option 3) in the STASH initialisation panel, and looking through your code I can't see any major differences from my own worked solutions, other than setting

       REAL, PARAMETER :: C_ALICE      = 1.0000  
       REAL, PARAMETER :: C_BOB        = 1.0000  

in ukca_constants.F90. You set these tracers to 1.0, which is different due to the number of zeros - this may affect the precision of the conversion from mass-mixing-ratio to volume-mixing-ratio (all the other conversion factors are to 4 decimal places). Could you change your ukca_constants.F90 to use 1.0000 and see what affect this has on your results?


comment:5 Changed 5 years ago by s1251469

That doesn't seem to have changed the output.

comment:6 Changed 5 years ago by luke

Could you fcm commit your changes to your branch - this allows me to take a checkout of it and make some changes in my copy of your job.


comment:7 Changed 5 years ago by s1251469


comment:8 Changed 5 years ago by luke

  • Resolution set to invalid
  • Status changed from accepted to closed

Were you able to get this working? If you are still having problems, you may find it easier to re-do the tutorials using the ARCHER versions:

I will close this ticket as it is HECToR specific - please re-open a new ticket for ARCHER work.

Many thanks,

Note: See TracTickets for help on using tickets.