questions about tidal forcing

[backkom]

Hi
here i wanna to make a simple test case in which model is only forced by tidal.i have some basic questions about tidal forcing. i have read some posts about this,but i am still not sure about these.
1.if the tidal forcing could be given in the boundary files? for example,my cases only has one south open boundary,can i impose the tidal forcing just by prescribing elevations and currents at south boundary without providing the tidal forcing file containing some harmonic constants of elevation and current.if only boundary files is possible,then i do not need the tidal forcing file,if not,why?
2.some posts said that both tidal elevations and currents on the open boundary should been given for tidal forcing case,i don not sure if this is true.if so ,why not simply providing the tidal elevations with normal currents at the open boundary set to zero.
3 why the tidal forcing should be given at the whole grid?it is seems that ROMS only uses the ones along open boundary.

i am new to roms and ocean model so above questions perhaps are quite basic.and i am looking forward to some reply or explain about this.
any reply will be appreciable!
thanks!

[kate]

1.if the tidal forcing could be given in the boundary files? for example,my cases only has one south open boundary,can i impose the tidal forcing just by prescribing elevations and currents at south boundary without providing the tidal forcing file containing some harmonic constants of elevation and current.if only boundary files is possible,then i do not need the tidal forcing file,if not,why?

Yes, you can specify the southern boundary condition as a function of time and not use a tidal forcing file.

2.some posts said that both tidal elevations and currents on the open boundary should been given for tidal forcing case,i don not sure if this is true.if so ,why not simply providing the tidal elevations with normal currents at the open boundary set to zero.

If you specify zero normal currents at the open boundary, it is not behaving like an open boundary. Tides going into an estuary will have variable elevation and currents, with maximum currents when the elevation is near zero, minimum currents at the extremes of the elevation.

3 why the tidal forcing should be given at the whole grid?it is seems that ROMS only uses the ones along open boundary.

This I don't know. However, since ROMS isn't using the interior, you can just set them to zero.

[backkom]

kate,thanks very much for your reply!
i still have another question about input file.if the tidal forcing file is provided,then what should my boundary file look like? for exmaple,only one south open boundary with onther three close boundary,if i have the forcing file,then the south boundary is not needed in boundary file,is is right?
thanks!

[backkom]

if i prescribe both harmonic constants in the forcing input file and also currents in the boundary file,for example,a
inflow along the open boundary, my question is weather or not the boundary currents should be the combination of the tidal forcing and inflow currents?
if so ,boundary currents and elevation in the boundary input file can be set to zero when tidal forcing file is set in a tidal forcing application.
thanks !

[kate]

You have choices:

1. You can specify a time-dependent boundary condition file that temporally resolves the tides and skip the tides file entirely.

2. You can specify the tides in the tidal file and skip the boundary condition file entirely - though this option might require you to tell ROMS in globaldefs.h that you really, really don't need an OBC file.

3. You can specify other 2-d currents and surface elevation signals as coming in, plus have tides and use the ADD_M2OBC and ADD_FSOBC #defines.

[backkom]

kate,thanks for your reply!

[arango]

I actually do not recommend to put the tidal forcing in the boundary file. Recall that the tides are specified spectrally in terms of frequencies. The boundary file contains time snapshots of the model state variables and may not resolve all the frequencies of the tides.

Now, the reason why we provide a full 2D grid of the tides is for analysis, comparisons and detiding purposes. During detiding we need to check the original tidal fields to see if the least-squares fit procedure is resolving all the frequencies over a particular time window.

[kate]

I know the FVCOM gang specifies all the tides in the boundary condition. The code they use to create the boundary condition knows to provide the 18-year tides as well as the higher frequencies. Of course, you'd want to compute the tidal boundary condition hourly to avoid aliasing.

[backkom]

hi,arango and kate
thanks for your suggestions! but i don not understand why not recommending to put the tidal forcing in the boundary file.for most applications the time interval between barotropic or baroclinic steps are several seconds or minutes,which are far less than the periods of tidal components,so i think this time snapshots are enough to cover all frequencies of tides. is it right?

[backkom]

Now, the reason why we provide a full 2D grid of the tides is for analysis, comparisons and detiding purposes. During detiding we need to check the original tidal fields to see if the least-squares fit procedure is resolving all the frequencies over a particular time window.

and arango,I don't catch your meaning,could your speak more clearly,please.i think the tides along the boundary are enough to forcing the model,is it right? are the full grid of the tides just for the comparisons with the results of least-squares fit?
thanks.

[kate]

It all depends on what you are planning to do. I plan to do multi-decadal simulations on the Northeast Pacific domain. I do not want to store the tidal boundary conditions on an hourly-basis, so will be using the tidal forcing file. If you are computing an estuary and only planning to run for a few weeks, then there is no reason not to force the tides through the boundary conditions. As for Hernan's point, are you going to detide your results or save the output with tides included? I need output for a short time with tides (hourly) as well as daily detided output as well as three-hourly output of the surface fields only. It's a challenge for one model to do it all.

[backkom]

It all depends on what you are planning to do. I plan to do multi-decadal simulations on the Northeast Pacific domain. I do not want to store the tidal boundary conditions on an hourly-basis, so will be using the tidal forcing file. If you are computing an estuary and only planning to run for a few weeks, then there is no reason not to force the tides through the boundary conditions. As for Hernan's point, are you going to detide your results or save the output with tides included? I need output for a short time with tides (hourly) as well as daily detided output as well as three-hourly output of the surface fields only. It's a challenge for one model to do it all.

kate,thanks for your explanations!
according to your words,the tidal forcing can both be given through boundary file or forcing file,the both methods have no difference in dynamics or physics. and for a long term simulation providing tides through the boundary file will result in a very large input file which is not convinent for our use ,then forcing file is a better choice at this time. when short term application both will be convinent and ok, am i right??
thanks !

[kate]

Yes, that is right. FVCOM is designed by and for people doing short-term estuary simulations which is why they don't provide the forcing file option, only the boundary file option.

[backkom]

Hi,kate
here thanks so much for your help!with your help i think i have understand this!
thanks!