Opened 4 months ago

Closed 3 months ago

#2540 closed task (fixed)

OSTIA SST and seaice fields

Reported by: anmcr Owned by: um_support
Priority: normal Component: Data
Keywords: OSTIA Cc:
Platform: Monsoon2 UM Version: 10.9

Description

Dear Willie,

I need to run the regional configuration of the UM from 1/7/2014 to 31/10/2014, on MONSooN. I already have the necessary 6-hourly ERA-Interim fields. But could you please also supply me with the necessary OSTIA seaice and SST fields.

Many thanks for your help.

Andrew

Attachments (5)

EC_grib.png (26.1 KB) - added by anmcr 4 months ago.
for-stu.PNG (81.5 KB) - added by anmcr 4 months ago.
ERA-Interim fields already retrieved
part.gif (18.7 KB) - added by willie 4 months ago.
Running the UM with Start Dumps built from ECMWF Grib Data - BAS Wiki.docx (25.7 KB) - added by anmcr 4 months ago.
for willie.PNG (197.1 KB) - added by anmcr 4 months ago.

Download all attachments as: .zip

Change History (24)

comment:1 Changed 4 months ago by willie

Hi Andrew,

I can get ERA Interim SST and sea ice from ECMWF in GRIB format - will that do? The OSTIA data is from the Met office and I'm not sure where it is.

Regards
Willie

Changed 4 months ago by anmcr

comment:2 Changed 4 months ago by anmcr

Hi Willie,

I've attached a screenshot of the 21 fields that Stu Webster has told me I need to run the nested suite. Are you able to retrieve these for the dates I require, from ERA-Interim?

Previously you retrieved 8 fields (skin temperature, surface pressure, land sea mask, geopotential, temperature, specific humidity, u velocity, v velocity).

Many thanks for your help,

Andrew

Changed 4 months ago by anmcr

ERA-Interim fields already retrieved

comment:3 Changed 4 months ago by anmcr

Hi Willie,

Could you please get me ERA-Interim SST and seaice for the period I asked for. You have already retrieved 8 fields for me from ERA Interim. See attached xconv screenshot. Could you please re-retrieve these fields as well, so that all the fields are on a single file?

Many thanks for your help.

Andrew

Changed 4 months ago by willie

comment:4 Changed 4 months ago by willie

Hi Andrew,

Do you really need these 21 fields? The UM will start with the basic eight and I can get most of the others. The ones that are not available are the cloud data fields. These may have been merged in from another program. I can get 18 of the 21 - see the attached picture part.gif.

Note that I don't have logarithm of surface pressure, but I do have surface pressure. For some reason in ERA-I the logarithm of the surface pressure is a multi level field! Perhaps Stuart just used the first level.

You need to think how the extra fields will be merged with the existing eight. One possibility is not to merge at all but for me to get the files like the one in my picture, ready to start the UM.

Let me know how you want to proceed.

regards
Willie

comment:5 Changed 4 months ago by anmcr

Hi Willie,

I discussed this with Stu yesterday, and have got his lastest version to run with the basic eight fields you sent me. However, would it be possible to add SST and seaice to the basic eight fields? See the comment I sent yesterday.

Many thanks,

Andrew

comment:6 Changed 4 months ago by anmcr

Hi Willie,

We're ready to run the model for the production run now, and only need the SST & seaice fields in addition to the basic eight fields (as discussed in the previous message). Are you please able to retrive these over the next day or so?

Many thanks,

Andrew

comment:7 Changed 4 months ago by willie

Hi Andrew,

Your data is now ready on Monsoon in /projects/nexcs-n02/frmy/data_xfer/Tkt2540_seaice_sst.grib. Please copy it to your own work space as soon as possible.

I have another user who is interested in running a regional model initialised from GRIB data. If you think this might be a good starting point for him, could you let me know the suite id.

Sorry it has taken so long, there was a heavy load on the ECMWF computers.

Regards
Willie

comment:8 Changed 4 months ago by anmcr

Dear Willie,

I discussed this with Stu Webster. There are a coupled of problems: i) The fields have to be merged into a single file for each analysis time, ii) The fields have to be at the same resolution (the SST/seaice fields are 512x256 grid points, while the others are 450x241). I don't know how to merge grib files.

Stu asked if you could please re-retrieve the fields into a single file per analysis time?

Regarding your question, I took a copy of Stu Webster's nested suite u-av356. Follow the instructions in the attached MS Word document (written by James Pope) to get it to work.

Many thanks,

Andrew

comment:9 Changed 4 months ago by anmcr

Willie,

Stu also sent me the message:

Yes, point him at https://code.metoffice.gov.uk/trac/rmed/wiki/suites/nesting (which I’ve just updated), and this does point to u-av356.

Best wishes,

Andrew

comment:10 Changed 4 months ago by willie

Hi Andrew,

I'm in the process of getting you 492 files at 6 hourly intervals containing only sea ice and sst at 450x241.

Thanks for the info on the suite

Regards
Willie

comment:11 Changed 4 months ago by anmcr

Dear Willie,

I have to add a lot of diagnostics to the STASH, some of which I am not familiar with. Is there a job that you could recommend that I could follow as a template?

Many thanks,

Andrew

comment:12 Changed 4 months ago by willie

Hi Andrew,

Your data request is still queued at ECWMF.

Adding STASH is reasonably simple if you are just adding new fields and using existing domains. See the discussion at http://cms.ncas.ac.uk/wiki/UmTraining/April2018. The STASH discussion starts at slide 116 of the presentation and more details can be found in section 6,3 of the practical exercises. Many people forget to run the stashindices.TidyStashTransform? macro after making changes, and this causes all sorts of problems.

Regards
Willie

Changed 4 months ago by anmcr

comment:13 Changed 4 months ago by anmcr

Dear Willie,

Thanks for the info. I am a little unsure how to set un the time profile to output accumulated amounts. I have attached a screen shot of my setup, which I have used before and which I think is correct. However, I was comparing it with one of Simon Vosper's jobs, which had a similar setup to mine with the exception that the variable 'intv' (specify the Time Interval for processing) was set to -1, and not 1 as I have. Are you able to advise what 'intv' should be set to?

Many thanks,

Andrew

comment:14 Changed 4 months ago by willie

Hi Andrew,

It looks OK to me. Your set up just means that the time step values will be accumulated for a period of one hour. Further details on the control of means is in UMDP C05.

Regards
Willie

comment:15 Changed 4 months ago by anmcr

Hi Willie,

Many thanks for the reply.

I actually want to accumulate for a period of 3 hrs, so presumably I therefore should have set intv=3.

Andrew

comment:16 Changed 4 months ago by willie

Yes, that's the idea.

comment:17 Changed 4 months ago by willie

Hi Andrew,

Your data is now ready in /projects/nexcs-n02/frmy/data_xfer on Monsoon. Please copy it to your own work space as soon as possible.

Regards
Willie

comment:18 Changed 4 months ago by anmcr

Dear Willie,

Thanks for retrieving the files. I have now copied them over.

I still have some difficulties finding all the STASH items I require. I have been using http://puma.nerc.ac.uk/STASH_to_CF/STASH_to_CF.html. I might need some help. Would you like me to start a new ticket?

Best wishes,

Andrew

comment:19 Changed 3 months ago by willie

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

Hi Andrew,

Yes, starting a new ticket would be good - could you explain there "I still have some difficulties finding …". Thanks.

I'll close this ticket now.

Regards
Willie

Note: See TracTickets for help on using tickets.