Opened 11 years ago

Closed 11 years ago

#250 closed help (fixed)

n320 global run with ECMWF start data

Reported by: anmcr Owned by: um_support
Component: UM Model Keywords: n320, ecmwf
Cc: Platform:
UM Version: 6.1



I am having difficulty running a N320 global run from reconfigured ECMWF start data. I would be grateful for some help.

So far I have
1) Reconfigured the ECMWF ERA-40 start data (job xdvha). This appears to be successful. The start data is in ~anmcr/work/UM_DATA/DAILY/SWE_Paper/1995071400_N320.astart.

2) I used a separate job to run the model with this start data (job xdvhb). This was based on the Hector PUM 6.1 template for N320. However, this run fails when running the model. I am not able to see an obvious error. I am confused as the NCAS-CMS technical report 1 states that once reconfigured the ECMWF start file can be treated as a normal MetO start file.

3) I have run the Hector PUM 6.1 N320 example job without any difficulty (job xdvhz).

4) Previously I was successful in running the N216 global model using ECMWF start files (xdpxf)

Thanks for any help. I've spend quite a while trying to sort this myself, but I've exhausted most of the possibilities I thought of.

Andrew (anmcr)

Change History (2)

comment:1 Changed 11 years ago by anmcr


I've managed to sort this out myself, so this ticket can be closed.

The problem was that I had reconfigured the ECMWF start dump with daily seaice and sst ancillaries, created using xancil. There was nothing wrong with the procedure. However, the daily seaice data that I used had a small 'hole' over the north pole, which was naturally transferred to the ancillary. Although the resulting reconfiguration appeared to work fine, when the model was run it failed. Filling this 'hole' resulted in a successful model run.

I hope no-one has had to waste their time over this ticket.

Best wishes,


comment:2 Changed 11 years ago by willie

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

Hi Andrew,

I think I got xdvhb to work, by switching on reconfiguration and selecting vertical extrapolation off (see below). It ran to completion without error. Anyway, I am glad you've sorted it out.



Job xdwla Title Global run 640x481x50 for Antarctica700x532
Job xdwlb Title a with recon and no vertical extrap
Difference in window atmos_Domain_Vert

→ Model Selection

→ Atmosphere

→ Model Resolution and Domain

→ Vertical

Radio button: Type of vertical interpolation

Job xdwla: Entry is set to 'Linear'
Job xdwlb: Entry is set to 'Linear with no extrapolation'

Difference in window atmos_InFiles_Start

→ Model Selection

→ Atmosphere

→ Ancillary and input data files

→ Start dump

Check box: Using the reconfiguration

Job xdwla: Entry is set to 'OFF'
Job xdwlb: Entry is set to 'ON'

Note: See TracTickets for help on using tickets.