imposing tidal boundary conditions in parent and child 1 way

[jivica]

Dear all,
in the past months there was never-ending story about tides, tidal dynamics BC inside ROMS model, hence think that some of you could benefit from paper we have recently published. Analysis of tidal dynamics as a function of BC that you use in a constellation of parent-child (or even only parent case) of 1 way nesting system. We used ROMS model in all out simulations from a simple ana case with rectangular basin and sea-mount case, progressively adding more flavor including internal tides, mesoscale interaction, ending in a complex system like Hawai'i. Energy, velocity, zeta,... deficits were estimated if you use "standard BC" in time with different frequency of BC information applied in the model ( 20 min, 1, 2 of 3 hours) or tidal spectral forcing via "tide.nc" approach.

For those interested having problem in getting the paper online I can send it as pdf via email.

Cheers,
Ivica


bib:
Article title: Analysis of Imposing Tidal Dynamics to Nested Numerical Models
Journal title: Continental Shelf Research
Corresponding author: Dr. Ivica Janekovic
First author: Dr. Ivica Janekovic
Final version published online: 10-JAN-2012
Full bibliographic details: Continental Shelf Research (2012), pp. 30-40
DOI information: 10.1016/j.csr.2011.11.017

[johnluick]

Good article. I have been using your method "1"; now I will try your method "2".

Have you written a script to do the tidal analysis and create the child BCs? If so I (and probably others) would appreciate a copy. Otherwise I'll write something that fits in with the new boundary condition scripts Arango just released. (I assume they don't already do it.) You mention t_tide...a great piece of software, but this job calls for some "bespoke" tools.

[jivica]

Well I have to say that I do not use the latest version of ROMS (using version 3.4 from Aug 2011)so I am doing all in a classical 1-way nesting. For that I found the easiest and by far fastest and memory efficient is to use "station file" approach. I create child grids (as much as I want) and then put open boundary locations (in some order) into station file. In that way parent is saving small but sufficient subset of data only along the needed boundaries more frequent than say in his or avg files. Once you have those files than it is easy to create boundary conditions for nests and even to create tide.nc files for nests (option 2) if having long enough time series. If you do not have strong internal tides or other complex interactions than you can do that once and use it for all later runs but in the Hawai'i region one have to be careful (interactions of mesoscale and tides, random phases, ...)

So short to point:
1) create nest grids
2) save boundaries into station file (I have matlab script to do that)
3) run parent using STATION option in .h
4) create pure brys in time from station.nc file for each of child grids (I have matlab script to do that)
or 4) create tide.nc files for child along with detided brys used as BC (I have matlab script to do that)

I can provide matlab functions that I have.

Cheers,
Ivica

[johnluick]

Hi Ivica,

thanks for the explanation and offer of scripts. I could use all the ones you mentioned. Not sure of the best way but one way is to email them to john.luick@sa.gov.au.

John

[hajji]

Well I have to say that I do not use the latest version of ROMS (using version 3.4 from Aug 2011)so I am doing all in a classical 1-way nesting. For that I found the easiest and by far fastest and memory efficient is to use "station file" approach. I create child grids (as much as I want) and then put open boundary locations (in some order) into station file. In that way parent is saving small but sufficient subset of data only along the needed boundaries more frequent than say in his or avg files. Once you have those files than it is easy to create boundary conditions for nests and even to create tide.nc files for nests (option 2) if having long enough time series. If you do not have strong internal tides or other complex interactions than you can do that once and use it for all later runs but in the Hawai'i region one have to be careful (interactions of mesoscale and tides, random phases, ...)

So short to point:
1) create nest grids
2) save boundaries into station file (I have matlab script to do that)
3) run parent using STATION option in .h
4) create pure brys in time from station.nc file for each of child grids (I have matlab script to do that)
or 4) create tide.nc files for child along with detided brys used as BC (I have matlab script to do that)

I can provide matlab functions that I have.

Cheers,
Ivica

Hi Ivica,

I am looking to run ROMS with nested grid. I will appreciate getting your matlab scripts.
thanks,
Hafedh (hafedh.hajji@globocean.fr)

[fdaryabor]

Dear Dr
Hi

For tide simulation using ROMS (one-way nested domain) used TPXO.6 (25 km) as boundary condition for the parent domain (with 50 Km resolution), so that the parent provides boundary condition for the child ( with 9 Km resolution). Compare RMS (amplitude and phase) of the parent, child and TPXO.6 with tidal gauge shows low RMS for the parent in comparison with TPXO.6 and also improving of RMS for the child compared with those. My reason for reviewer is bathymetry errors could improved for the parent in comparison TPXO.6, so more accuracy for the child.

Could you please give me your comment or suggestion about above issue via my email, that is easy for me.

Beside could you please send your paper to me via my email.

farshiddaryabor@gmail.com

Cheers
Dr. Farshid Daryabor