Velocity is too large near open boundary

General scientific issues regarding ROMS

Moderators: arango, robertson

Post Reply
Message
Author
hezhi
Posts: 10
Joined: Wed Apr 11, 2018 2:31 pm
Location: FIO

Velocity is too large near open boundary

#1 Unread post by hezhi »

ROMS Users:
I find the velocity is too large near open boundary, but it don't blow up。I try to change .h file,but it does not work。What can I do for next step?
Attachments
roms_Indian_Ocean.in
(132.09 KiB) Downloaded 181 times
indian_ocean.h
(1.94 KiB) Downloaded 180 times
vel_gif.gif
vel_gif.gif (1.44 MiB) Viewed 8320 times

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: Velocity is too large near open boundary

#2 Unread post by kate »

You haven't told us anything about your boundary conditions except which options you have. What is the frequency of providing new information and what is the source of it? I remember seeing a vast improvement when going from monthly climatology of just T and S to 5-daily SODA fields. The climatology is a collection of whatever measurements they have for a given season/location, not at all consistent with each other, while the SODA fields are internally consistent views of T, S and velocity, also SSH. Daily is better than monthly, if you can get it.

hezhi
Posts: 10
Joined: Wed Apr 11, 2018 2:31 pm
Location: FIO

Re: Velocity is too large near open boundary

#3 Unread post by hezhi »

Dear Kate
the frequency of boundary fields are monthly,which is from monthly data of ORAS4. ERA5 data is source of my force fields. The data don't belong to climatology。
Last edited by hezhi on Wed Dec 23, 2020 6:25 am, edited 1 time in total.

hezhi
Posts: 10
Joined: Wed Apr 11, 2018 2:31 pm
Location: FIO

Re: Velocity is too large near open boundary

#4 Unread post by hezhi »

Could you recommend some kinds of data are better for running a long time(from 1978-12 to 2015-12)?and I don't kown if my options is right?

sissalVE
Posts: 9
Joined: Wed Feb 21, 2018 2:08 am
Location: University of Copenhagen

Re: Velocity is too large near open boundary

#5 Unread post by sissalVE »

Hi,
Could you show your bathymetry of both sides of the open boundary and your grid?

hezhi
Posts: 10
Joined: Wed Apr 11, 2018 2:31 pm
Location: FIO

Re: Velocity is too large near open boundary

#6 Unread post by hezhi »

Those are my bathymetry.
Attachments
d144bb07b2ad7813e01c45496842ccf.png
d144bb07b2ad7813e01c45496842ccf.png (552.59 KiB) Viewed 8206 times
Last edited by hezhi on Wed Dec 23, 2020 12:17 pm, edited 2 times in total.

hezhi
Posts: 10
Joined: Wed Apr 11, 2018 2:31 pm
Location: FIO

Re: Velocity is too large near open boundary

#7 Unread post by hezhi »

These are bathymetry of open boundary.
Attachments
006bc2a1e9d937b480aeab0cbd7dc58.png

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: Velocity is too large near open boundary

#8 Unread post by kate »

There are reanalyses out there at better than monthly, starting before 1979. You might have to contact the authors and mail them some harddrives to load it onto to get it, though. A colleague of mine has done that to get both SODA and Mercator.

sissalVE
Posts: 9
Joined: Wed Feb 21, 2018 2:08 am
Location: University of Copenhagen

Re: Velocity is too large near open boundary

#9 Unread post by sissalVE »

To me it looks like the issue is at the northern boundary. You could try to either expand the grid to start further north at -5, or so, depending on the bathymetry. So that the bathymetry at the boundary is less steep. Or reduce the Eastern ( or is it really the Western?) boundary to 60 degrees instead of 50.

Post Reply