Bug report of d_ecmwf2roms.m in surface forcing generation script

Bug reports, work arounds and fixes

Moderators: arango, robertson

Post Reply
Message
Author
asidjazz
Posts: 3
Joined: Mon Feb 06, 2023 4:01 pm
Location: Hong Kong Baptist University

Bug report of d_ecmwf2roms.m in surface forcing generation script

#1 Unread post by asidjazz »

Hello,

When I set "doFields = [1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16]", it generates forcing file with inaccurate values, especially for swflux and rain. The scale are about 20000 times higher and there are value < 0.

When I set doFields to a single number to generate a single file each time with "clear all" command in between, the files look good. I used absolutely the same input data.

I found it because the model kept blowing up, and super high fresh water flux was found in the middle of the ocean :shock:

Now I added a loop to change the single number of doFields, and "clear all" at the end of each loop. It works fine, but I still wonder why. Maybe something wrong with my parameter setting in the script or in matlab?

System: Linux mint 24.1
Matlab: 2021b

The d_ecmwf2roms.m is attached. Please check. Thank you so much.
Attachments
d_ecmwf2roms1.m
(18.29 KiB) Downloaded 12 times
Last edited by asidjazz on Tue May 13, 2025 9:59 am, edited 1 time in total.

asidjazz
Posts: 3
Joined: Mon Feb 06, 2023 4:01 pm
Location: Hong Kong Baptist University

Re: Bug report of d_ecmwf2roms.m in surface forcing generation script

#2 Unread post by asidjazz »

Also, the the swflux and rain, the files only have values at frame 1, and all 0 afterwards. swflux and rain files only. The others have values at all the frames.

Post Reply