HYCOM water level: 2023 GLBy0.08 vs 2024–2025 ESPC-D-V02 — why so different?

16 views

Skip to first unread message

Fatemeh Ameri

unread,

Nov 24, 2025, 8:38:59 PM (4 days ago) Nov 24

to HYCOM.org Forum

Hi all—looking for hourly water level for D-Flow FM boundaries near Cape Fear / Frying Pan Shoals (27–35°N, 81–69°W).

2024–2025: ESPC-D-V02/ssh/<year> → surf_el looks nicely tidal and hourly.
- e.g. https://tds.hycom.org/thredds/dodsC/ESPC-D-V02/ssh/2025
2023: GLBy0.08/expt_93.0/ssh/2023 → surf_el over the same box/time looks much less harmonic, smaller variance.
- e.g. https://ncss.hycom.org/thredds/ncss/GLBy0.08/expt_93.0/ssh/2023?var=surf_el&accept=netcdf4

Questions:

Are these surf_el variables physically equivalent across products, or is one de-tided/MDT-referenced?
Is there a 2023 archive with tidal sea level consistent with ESPC-D-V02 (hourly), or should I add astronomic tide separately in the model?
Any recommended path/endpoint for 2023 that matches ESPC-D-V02 behavior?

Thanks!

Alan Wallcraft

unread,

Nov 25, 2025, 6:35:12 AM (3 days ago) Nov 25

to HYCOM.org Forum, Fatemeh Ameri

GOFS 3.1 does not include tidal forcing and also does not include surface atmospheric pressure forcing. Both are included in ESPC-D-V02.

See Tidal Analysis of ESPC-D-V02

The ssh and steric_ssh from hourly "sur" files from GOFS 3.1 are the equivalent of hourly surf_el and steric_ssh from ESPC-D-V02. The steric SSH effectively removes external tides and so should be directly comparable. Or you could use the daily means from ESPC-D-V02, or you could subtract out the tides via the tidal analysis. Note that in general HYCOM only "cares" about gradients in SSH, and so two model runs could have a constant offset in space without this being significant.