Opened 7 years ago

Closed 7 years ago

#841 closed help (fixed)

n96L85 run fails to reconfigure

Reported by: anmcr Owned by: um_support
Component: UM Model Keywords:
Cc: Platform:
UM Version: 7.8

Description

Hi,

I am trying to run a N96L85 model which was recommended to me by Luke Abraham at Cambridge University — the GA3.0 HadGEM3-A model, xgvha.

My copy of this, xhagi, fails during the reconfiguration stage with the error

*
ERROR!!! in reconfiguration in routine Rcf_Calc_Len_Ancil
Error Code:- 1
Error message:- Stashcode is not a valid prognostic variable, Stash Code 121
Error generated from processor 0
*

There was a previous ticket with this error but it didn't give me any additional information. The startdump I am using is from a N512L70 job (xhagr) after having run for 1 hour. So far the only change I have made is to switch off the 'aerosol' physics as the reconfiguration was complaining that S02 or something was not in the startdump.

I would be very grateful of any assistance in getting this run to work.

Thanks,

Andrew

Change History (9)

comment:1 Changed 7 years ago by willie

Hi Andrew,

STASH 121 is 3D NATURAL SO2 EMISSIONS KG/M2/S. This isn't in the dump xhagga_da001 and you've switched off the sulphur cycle. I think you should set the sulphur cycle emissions(3D) ancillary to "not used" and try again.

Regards,

Willie

comment:2 Changed 7 years ago by anmcr

Hi Willie,

I did as you suggested, and I now get the same type of error during reconfiguration but for 418

'Stashcode is not a valid prognostic varaible, Stash Code 418'

I couldn't find #418 as an entry for a stash code. Also, I'm very unfamiliar with this type of N96 run. Are you able to provide further help?

Thanks,

Andrew

comment:3 Changed 7 years ago by willie

Hi Andrew,

The UMUI has gotten a bit confused. You need to

  • Go to Scientific Section 17 Aerosol
  • Tick scheme 2A
  • Press DUST and the select ancillary not used and close
  • Tick scheme 0A to put back "no aerosol"

It should then work.

Regards,

Willie

comment:4 Changed 7 years ago by anmcr

Hi Willie,

Thanks for your help.

It gets a bit further on with the reconfiguration but still fails. The error is


ERROR!!! in reconfiguration in routine Rcf_Set_Data_Source
Error Code:- 30
Error Message:- Section 0 Item 152 : Required field is not in input dump!

Firstly, I'm not sure where to lookup these stash codes. On the internet item 152 seems to refer to 'orographic gradient XX component' — but this is included in the qrparm.orog file and anyway not required.

Andrew

comment:5 Changed 7 years ago by willie

Hi Andrew,

I have written a little tool called check_stash: it is in /homee/n02/n02/wmcginty/bin. It takes two arguments: the stash number (up to three digits) and the version number:

check_stash 152 7.8
1| 1 | 0 | 152 |RIVER DIRECTION |
1| 1 | 4 | 152 |BULK CLOUD VOL INCR: pc2 checks |
1| 1 | 5 | 152 |QCL INCR: pc2 turbulence negative |

So if you look in the River routing global model ancillary, you'll find that you are configuring the river sequence: set this to not used. And perhaps switch off river routing if you're not using it.

Regards,

Willie

comment:6 Changed 7 years ago by anmcr

Hi Willie,

I did as you suggested (switched off River routing ancillary and switched off River Routing) which has moved the reconfiguration along a little. However, it's still failing. This time I can't even see what the error is.

/work/n02/n02/anmcr/xhagi/bin/qsexecute: Error in dump reconfiguration - see OUTPUT

but OUTPUT doesn't seem to give any additional information!

Thanks for all your help,

Andrew

comment:7 Changed 7 years ago by willie

Hi Andrew,

This time you need to look in the output in the work directory. Here it is complaining about stash 274 MEAN TOPOGRAPHIC INDEX. This is needed because you have large scale hydrology switched on. I'm not sure how to resolve this. I began by switching JULES off (in sub model choices) and then going to the Large scale hydrology ancillary page, push HYDROL selecting surface hydrology not included. But this gives an inconsistent set up. On the HYDROL page if you push LSH the ancillary file $ANCIL_LAND/topog_index_mean_polfix is the source of the STASH.

I hope that helps,

Regards,

Willie

comment:8 Changed 7 years ago by anmcr

Hi Willie,

I got it to reconfigure in the end and the job ran perfectly after that.

Thanks for all your help,

Andrew

comment:9 Changed 7 years ago by willie

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