Opened 3 months ago

Closed 6 weeks ago

#3121 closed help (fixed)

Change start dump time

Reported by: mtodt Owned by: um_support
Component: UM Model Keywords: time start dump ancillary
Cc: Platform: ARCHER
UM Version: 11.0

Description

Hi

I'm trying to run a Met UM suite from 1979 onwards using a start dump for 1982. I've set the model base time in rose-suite.conf to 1979 and i_override_date_time=2 and new_date_time=1979,1,1,0,0,0 in app/um/rose-app.conf. However, atmos_main crashes with the following error:

????????????????????????????????????????????????????????????????????????????????
???!!!???!!!???!!!???!!!???!!!       ERROR        ???!!!???!!!???!!!???!!!???!!!
?  Error code: 402
?  Error from routine: UP_ANCIL
?  Error message: REPLANCA: Current time precedes start time of data
?  Error from processor: 123
?  Error number: 93
????????????????????????????????????????????????????????????????????????????????

I have also set ancil_reftime=1978,12,1,0,0,0 so the simulation time step shouldn't precede the ancillary file. The .out log file lists prognostic fields with validity times of 1982:

Prognostic fields

Key to output - dump or other
Data stored on land/sea points
No of fields
Length of field before unpacking (==LBLREC)
Data type
STASH Code
Start of first field
Description
Validity time (yyyy:mm:dd:hh:mn:ss)
Forecast period
0   85  139968   1     2     1 U COMPNT OF WIND AFTER TIMESTEP        1982: 1: 1: 0: 0: 0   0
0   85  140400   1     3    86 V COMPNT OF WIND AFTER TIMESTEP        1982: 1: 1: 0: 0: 0   0
...

Does that mean resetting the date times fails or is there anything I have to do additionally? Thanks a lot in advance!

Cheers
Markus

Change History (7)

comment:1 Changed 3 months ago by grenville

Markus

It appears that data in some of the ancillary files has dates which are not compatible with a 1982 start date.

What suite id does this error occur in?

Grenville

comment:2 Changed 3 months ago by mtodt

Hi Grenville

The suite id is u-bp248.

But the suite runs when the start date is 1 January 1982 and the ancillary reftime is 1 December 1981, so do you mean that some ancillary files don't work when (p)re-dating the simulation to 1 January 1979 and ancillary files to 1 December 1978?

Cheers
Markus

comment:3 Changed 3 months ago by grenville

Markus

u-bp248 is failing in reconfiguration - it's hard to know where to look while the suite is changing.

Grenville

comment:4 Changed 3 months ago by mtodt

Hi Grenville

Sorry about that, I had to use this suite for a test as a copy of it failed. I've changed it back and the same error mentioned above pops up again. Thanks for your help!

Cheers
Markus

Last edited 3 months ago by mtodt (previous) (diff)

comment:5 Changed 3 months ago by grenville

Markus

Your model updates fields from ancillary files - it's complaining about /work/y07/y07/umshared/ancil/atmos/n216e/orca025/seaice/reynolds/1981_2012_360/v2/qrclim.seaice, because the the first valid data in that file is 1981/09/01 — later that the model start time.

Grenville

comment:6 Changed 3 months ago by mtodt

Thanks, Grenville!

But the first field is /work/y07/y07/umshared/ancil/atmos/n216e/ozone/sparc/1994-2005/v3/qrclim.ozone_L85_O85. Shouldn't that error occur for this file as well?

Cheers
Markus

comment:7 Changed 6 weeks ago by grenville

  • Resolution set to fixed
  • Status changed from new to closed

Markus

That ozone file doesn't suffer the same problem

Grenville

Note: See TracTickets for help on using tickets.