search

ACCESS-NRI / accessdev-Trac-archive

Archive accessdev Trac contents as issues
Apache License 2.0
0 stars 0 forks source link

Failure of AMSU-A Channel 15 on Metop-B #291

Closed penguian closed 7 years ago

penguian commented 7 years ago

keyword_AMSU-A_ATOVS_channel_gross_limits_check_channel_selection resolution_fixed | by jtl548@nci.org.au

Channel 15 of AMSU-A onboard Metop-B failed around 01Z on 17th October, 2016. Here's a message posted to the ITWG mailing list by Fiona Smith about the failure,

From: owner-itwg_nwp@metoffice.gov.uk [mailto:owner-itwg_nwp@metoffice.gov.uk] On Behalf Of Smith, Fiona Sent: Tuesday, 18 October 2016 7:52 PM To: itwg_nwp Subject: Metop-B AMSU-A Channel 15

The EARS monitoring that we run has identified a problem with Metop-B AMSU 15, which seems to have failed at around 01Z on 17th October. The warm target counts and space counts both became equal. We're not using it operationally, but have removed it from our parallel suite, and the gross limits check we apply has meant that a lot of data is being rejected. We're adjusting the gross limits check also so Met Office monitoring for other channels should return to normal over the next day or two.

Fiona

No more information at the present time.

Please note: I work part-time and may not be in on a Monday

-- Fiona Smith (née Hilton) Expert Scientist Met Office FitzRoy Road Exeter EX1 3PB United Kingdom Tel: +44 (0)1392 886431 Fax: +44 (0)1392 885681 Email: fiona.smith@metoffice.gov.uk http://www.metoffice.gov.uk

Find out more about our world leading science and partnerships at http://www.metoffice.gov.uk/research

The research APS2 global suite available at NCI does not use channel 15. However, as indicated in the email from Fiona because of the failure of gross limits check by this channel all the channels - even good ones - from Metop-B are rejected.

This ticket describes an update to ATOVS channel selection file to switch off the gross limits check on channel 15. Note that this update and test are only applicable to the research APS2 global suite.

Issue migrated from trac:291 at 2024-01-31 18:26:14 +1100

penguian commented 7 years ago

@jin.lee@bom.gov.au changed status from new to assigned

penguian commented 7 years ago

@jin.lee@bom.gov.au set owner to jtl548

penguian commented 7 years ago

@jin.lee@bom.gov.au commented

I used a suite, au-aa551 which is a copy of the research APS2 global suite, au-aa427.

Following set-up was used:

N.B. Because the suite is old it is not compatible with the default Rose and Cylc on accessdev. I used older versions of Rose and Cylc by using, export CYLC_VERSION=6.4.1; export ROSE_VERSION=2015.04.1

  1. The control used the default channel selection file, raijin1:/projects/access/da/nwp_input/ops_var/Data/SatRad_coeffs/bom/users/au-aa427/aps2_r00_my00.v114/ATOVS_channels.nl. The stats output log from this control run is https://accessdev.nci.org.au/rose-bush/view/jtl548/au-aa551?path=log/job/2015103112/Ops_atovs/04/job.GEN/stats

  2. Updated ATOVS channel selection file is raijin1:/projects/access/da/nwp_input/ops_var/Data/SatRad_coeffs/bom/users/au-aa551/aps2_r00_my00.v114/ATOVS_channels.nl (c.f. ATOVS_channels.nl.orig). The stats output log is https://accessdev.nci.org.au/rose-bush/view/jtl548/au-aa551?path=log/job/2015103112/Ops_atovs/06/job.GEN/stats

The stats output log files from both runs are identical.

I also tested the change in the channel selection file using a Bufr file containing data from faulty AMSU-A channel 15 from Metop-B. Following files were retrieved from sam archive (samsrv2:/samaccess_prod/ACCESS_prod/access_g),

  1. The stats output log from the control run with default channel selection file is https://accessdev.nci.org.au/rose-bush/view/jtl548/au-aa551?path=log/job/2016110312/Ops_atovs/01/job.GEN/stats. Here all the channels from Metop-B (labelled MetOp1 ATOVS) are rejected.

  2. The stats output log from the updated ATOVS channel selection file is https://accessdev.nci.org.au/rose-bush/view/jtl548/au-aa551?path=log/job/2016110312/Ops_atovs/02/job.GEN/stats. Here statistics for all the channels from Metop-B are written out as well as more observations are written out to varobs file.

penguian commented 7 years ago

@jin.lee@bom.gov.au changed _comment0 which not transferred by tractive

penguian commented 7 years ago

@jin.lee@bom.gov.au changed status from assigned to closed

penguian commented 7 years ago

@jin.lee@bom.gov.au set resolution to fixed