Opened 12 years ago

Closed 11 years ago

#118 closed error (fixed)

Negative theta detected

Reported by: rdamoah Owned by: lois
Component: UM Model Keywords:
Cc: Platform:
UM Version:

Description

Hi
I am doing a future run starting 2088 of UM with Stochem/slimcat shemes, but the run crashes after one month with the error ATM_DYN : NEGATIVE THETA DETECTED. The name of my job is xcusa.

I also have a present day job which is running ok without any problems. The difference between these two jobs are the start date, ancilliary files and the CO2 values.
Thanks

Change History (2)

comment:1 Changed 12 years ago by lois

  • Owner changed from um_support to lois
  • Status changed from new to assigned

Identifying reasons why your model run has blown up is one of the most difficult questions to answer.

You are using a a non-standard configuration (vn4.5 L64 with STOCHEM/Slimcat chemistry etc) which is stable enough to run in your first case. You then have introduced 3 changes which make it unstable. Without looking in detail at the changes you have made could I suggest

  • you check your new ancillary files as I don't know how you created them
  • you check the sensitivity of your model setup to each of your changes individually rather than all 3 at once
  • you look at the model output just before it blows up at timestep 1615. Sometimes just reconfiguring a restart dump (which tends to 'smooth' the data) and re-starting your model run can help it to keep going. Sometimes just reducing the timestep for the period where it has difficulties can also help keep it going. These sort of 'fixes' will only really work if the model results look 'okish' before it blows up, you will have to be a judge of what you think is the definition of 'okish'.

Please get back to us if you are still having problems after you have done some further investigations.

Lois

comment:2 Changed 11 years ago by ros

  • Resolution set to fixed
  • Status changed from assigned to closed
Note: See TracTickets for help on using tickets.