Time type in nc file

[ara952]

Hi
Could you please let me know what is the type of time for running the ROMs in the nc output file?
I mean, is it local or GMT?
Thank you

[kate]

It doesn't matter. Just be consistent across all your inputs.

[ara952]

I extracted the zeta variable to compare to the published tidal constituents from NOAA stations in the Gulf of Maine. The amplitude and phase from the model output are very different from the NOAA ones. I picked one of the NOAA stations position and found the closest node from the model to that position and extracted the zeta variable and run t_tide. The M2 constituent for the model is 0.8 m while the NOAA one is 2.018 m. Could you help me with this problem?
Thank you

[arango]

ROMS has nothing to do with the time convention and elapsed time. The user selects how the time is managed in an application. There is nothing that we can do about it. It is up to the person(s) that set-up that application to use a consistent time zone in all the input NetCDF files (initial conditions, boundary conditions, atmospheric forcing, and tidal forcing). Usually, all the databases out there are in GMT and set the appropriate time units since the reference time (epoch).

[wilkin]

I extracted the zeta variable to compare to the published tidal constituents from NOAA stations in the Gulf of Maine. ...

Whose ROMS model output did you analyze?

[johnluick]

ara952, of course Hernan and Kate are right about the time, but if you use TIDE_START=-2 and initialise with GMT phases for a particular time and plan to analyse with t_tide (which assumes all harmonics GMT phase, same as TPXO and all the other global tidal datasets that I know of) you need to think strictly in terms of GMT. I don't use ROMS much any more and haven't thought about this much in a while plus my brain is ossified but I have some notes at the start of my toolkit for setting up the tides input file (look at TPXO2ROMS_v4pt1_script.m which is on austides.com/downloads).

If you want to compare to harmonics phases in local time (post model run) let me know, I can explain that.