comparing the surf_el (surface elevation) variable between: GLBy0.08 expt_93.0 and ESPC-D-V02

[Márcio Yamashita]

Hello HYCOM users,

I have been trying to create a time series of surf_el using GLBy0.08 expt_93.0

(decommissioned last September) and ESPC-D-V02: Global 1/12° Analysis (Aug-10-2024 to Present with 8-day forecast).

GOFS 3.1 does not have tides,  in ESPC-D tides are apparent. I cant merge surf_el of both models.

Another doubt

Is ESPC-D contains  tides currents added in U and V currents components?

[Alan Wallcraft]

See Clarification on Filtering Tides within ESPC-D product

You can use espc_mean25 to form a daily (running) mean of SSH.  Note that this can probably also be applied to hourly GOFS SSH.

Or you can use hourly steric SSH, which removes the external tide signal.

However, I]m not sure if either approach will produce compatible datasets.  An ocean model only cares about the gradient of SSH, so SSH can be offset by a spatially constant field with no effect.  So the two SSH data sets might be offset by a constant, or perhaps by a non-constant offset with weak gradients.

Alan.

[Alan Wallcraft]

ESPC-D-V02 includes surface air pressure forcing, so it effectively has an inverse barometer included in the surf_el.  This and the external tides are not present in steric SSH, which is definitely the fields to compare to GOFS 3.1.

Alan.