Opened 9 years ago

Closed 9 years ago

#491 closed help (fixed)

ancillary file service

Reported by: anmcr Owned by: grenville
Component: Data Keywords:
Cc: Platform:
UM Version: 7.1

Description

Hi,

I used the NCAS ancillary service retrieve the landmask and orography ancillaries for a 4 km LAM domain over a section of Antarctica. However, the landmask and orography files do not seem correct as they indicate all sea points and zero elevation.

The coordinates I entered were

no. of cols: 750
no. of rows: 750
col spacing: 0.036
row spacing: 0.036
1st lat: -48.87
1st lon: 187.52
pole lat: 21.73
pole lon: 8.07

and the resulting ancillaries are on puma under Durmont….

When I plot the domain using IDL and my own orography dataset I get what appears to be the correct domain. See attachment. Which is 50/50 land/sea.

Thanks,

Andrew

Attachments (1)

dumont_orog.png (27.5 KB) - added by anmcr 9 years ago.
domain plotted using idl

Download all attachments as: .zip

Change History (7)

Changed 9 years ago by anmcr

domain plotted using idl

comment:1 Changed 9 years ago by anmcr

Hi again,

I did some more tests with different IGBP aggregation methods:

Method 2: correct qrparm.mask, wrong qrparm.orog

Method 5: wrong qrparm.orog, wrong qrparm.orog

By 'wrong' I mean the fields are all zero's.

Andrew

comment:2 Changed 9 years ago by grenville

  • Owner changed from willie to grenville
  • Status changed from new to accepted

Andrew

The log files don't indicate anything useful, ie the program didn't crash, and it thinks it has been successful. The fact that both land-sea mask and orography indicate the same thing might be an indication that it thinks the domain is really over the sea - could you give me the parameters of a domain (close to this one if possible) that was successful. I note that if you simply supply an unrotated domain at about the same place, the CAP seems to handle it OK.

Grenville

comment:3 Changed 9 years ago by anmcr

Hi Grenville,

A domain close to this which works is

Cols: 750
Rows: 750
dx: 0.036
dy: 0.036
1st lat: -46.58
1st lon: 169.85
Pole lat: 21.73
Pole lon: 8.07

This has the same latitude and longitude pole as the one which didn't work.

Maybe I could move my domain around a little, and that might work.

Thanks,

Andrew

comment:4 Changed 9 years ago by grenville

Andrew

I've been playing with the CAP on different domains - there does seem to be a problem for certain choices, but it ran OK for the following:

no. of cols: 750
no. of rows: 750
col spacing: 0.036
row spacing: 0.036
1st lat: -46.0
1st lon: 187.5
pole lat: 21.73
pole lon: 8.07

The data is in /export/puma/data-01/ancil/aorr.1284024263

More work needs to be done to find the source of the error - I shall forward the problem to the Met Office to see if they have the same issue (they won't be running this version of the CAP 'though), but you might be able to get a reasonable with some tweaking if its parameters or use this one.

Grenville

comment:5 Changed 9 years ago by anmcr

Dear Grenville,

Thanks for helping me with this.

Please close this ticket.

Andrew

comment:6 Changed 9 years ago by grenville

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