Tidal angular frequency has units [rad s-1]

[awallcraft]

Tidal frequencies are always angular frequencies to simplify applying sine and cosine. These have MKS units [rad s-1] but they are all currently listed as [s-1].

Updated dOxygen comments for variables, e.g. [T-1 ~> s-1] becomes [rad T-1 ~> rad s-1]. Updated get_param units. e.g. units="s-1" becomes units="rad s-1".

No answers are changed, but the logged parameter units are different.

There are frequencies in MOM_internal_tides.F90 but these have not been updated because they may be specified incorrectly. They are used as if they are [T-1] but they are calculated as 2PI/period [rad T-1].

real, allocatable, dimension(:) :: frequency !< The frequency of each band [T-1 ~> s-1].

real :: period ! A tidal period read from namelist [T ~> s]

! The periods of the tidal constituents for internal tides raytracing
call read_param(param_file, "TIDAL_PERIODS", periods)

do fr=1,num_freq
period = US%s_to_Textract_real(periods, " ,", fr, 0.)
CS%frequency(fr) = 8.0atan(1.0)/period
enddo

All MOM6-examples cases have INTERNAL_TIDES=False and so can't resolve this issue.

[Hallberg-NOAA]

These changes to the documented units seem sensible. We have been considering changing some of the sines and cosines to use arguments in degrees so that they have more exact periodicity, and if we were to do so, having the clear and explicit documentation of which frequencies are in radians per second that this PR introduces will be very helpful.

[Hallberg-NOAA]

This PR has passed pipeline testing at https://gitlab.gfdl.noaa.gov/ogrp/mom6ci/MOM6/-/pipelines/25707, with the expected changes in the documented units in MOM_parameter_doc files.