Fix history files out sync with model components

[ezhilsabareesh8]

Pull Request Summary

This PR fixes a WW3 output-time inconsistency where the timestamps in WW3 history NetCDF files drifted from the coupled model “true” time after restarts (e.g., monthly or annual restarts), leading to apparent time jumps when reading the history files.

Issue addressed: iss831

[anton-seaice]

Thanks @ezhilsabareesh8 - can you add test evidence please?

For example, a pre-release build of access-om3 and extracts from ncdump showing the new dates when running the new model would be appropriate.

[ezhilsabareesh8]

Thanks @anton-seaice , Here is the plot of model time index (in days) against the date times of the respective history files based on the draft PR ACCESS-NRI/ACCESS-OM3#170. The outputs are here /g/data/tm70/ek4684/771-dev-MC_100km_jra_ryf/archive.

[anton-seaice]

Looks good:

ncdump -v time access-om3.ww3.hi.1900-04-21-00000.nc
netcdf access-om3.ww3.hi.1900-04-21-00000 {
dimensions:
nx = 360 ;
ny = 300 ;
time = UNLIMITED ; // (1 currently)
freq = 25 ;
variables:
double time(time) ;
time:units = "seconds since 1900-01-01 00:00:00" ;
time:calendar = "noleap" ;
double lon(ny, nx) ;
...
data:

time = 9504000 ;
}

(110 days in noleap calendar)

[anton-seaice]

Looks good! Thanks :)