search

NOC-MSM / pyBDY

pyBDY: a Python based regional NEMO model configuration toolbox.
GNU General Public License v3.0
7 stars 7 forks source link

Pynemo crashes when wrapping coordinates in Orca #32

Open jenjar93 opened 4 years ago

jenjar93 commented 4 years ago

Pynemo crashes when attempting to wrap coordinates in Orca around 75 degrees E

jdha commented 4 years ago

Jenny - do you have an example I can test? I have access to the Liverpool filesystem (also JASMIN/ARCHER). Thanks James

jenjar93 commented 4 years ago

Hi James

Yep, I've moved some example files onto ARCHER in: /work/n01/n01/jenjar93/for_jdha.

Let me know if there are any permissions issues for this. The original files have been copied from my folder on livljobs

(/projectsa/accord/SANH_jenjar/OPEN_BOUNDARIES), but this was primarily a test directory and so it's quite messy.

The problem we were having was that the ORCA grid cuts directly in the middle of my model domain (India and adjacent seas). When we tried to run PyNEMO, it didn't want to wrap around the coordinates, and so failed. Ash previously got around this problem by manually cutting and splicing the ORCA domain before running PyNEMO.

Let me know if you need any more information.

Thanks!

Jenny

From: jdha notifications@github.com Sent: 06 December 2019 12:29:48 To: NOC-MSM/PyNEMO Cc: jenjar93; Author Subject: Re: [NOC-MSM/PyNEMO] Pynemo crashes when wrapping coordinates in Orca (#32)

Jenny - do you have an example I can test? I have access to the Liverpool filesystem (also JASMIN/ARCHER). Thanks James

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHubhttps://github.com/NOC-MSM/PyNEMO/issues/32?email_source=notifications&email_token=AM4P2SUZ2BBZI6IBNEDDI2TQXJATZA5CNFSM4I4XCEF2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEGD6DTQ#issuecomment-562553294, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AM4P2SR5ANVLRFVIDZHN2JDQXJATZANCNFSM4I4XCEFQ.

jdha commented 4 years ago

Hi Jenny,

Thanks for that - permissions look good. I think in the past I've relied on the fact that the global NEMO data has a single wrap point on the eastern and western extent (i.e. there is a repeated column at i=0 and I=-1) I've managed to use pynemo (in it's previous Matlab form) for creating a southern ocean model with a northern boundary that transects this discontinuity. So - I must have written a solution (albeit in Matlab) at some point. I'll delve into my old scripts and update the python (hopefully before the Christmas break).

James