Model [2025.12.31]¶
How to download widely used numerical model products: reanalyses, analyses and forecasts.
Before using model output
Even without configuring and running a numerical model yourself, good research is possible with the many distributed model products. Which model reproduces the real ocean best differs from region to region, however, so evaluating the product against observations is an essential step before any serious analysis. Many earlier studies have done this; look them up.
Personally I rely on HYCOM and GLORYS, which have high spatial and temporal resolution and assimilate observations. Neither GLORYS nor HYCOM includes tides, although the HYCOM ESPC product distributed since 2024-08-10 is reported to include them.
Global reanalysis & analysis¶
HYCOM¶
Data-assimilating global ocean model. The ESPC product includes tides from 2024-08-10 onward
How to download HYCOM (MATLAB: ESPC, GOFS 3.1, earlier versions)
Tidal Analysis of ESPC-D-V02 (HYCOM forum)
GLORYS12v1¶
0.083° · daily/monthly · 1993-01-01 – (almost) present
CMEMS global reanalysis (GLOBAL_MULTIYEAR_PHY_001_030)
GLO12v4¶
0.083° · hourly/daily/monthly · 2022-06-01 – forecast (+10 d)
Global Ocean Physics Analysis and Forecast. Short period, but high temporal resolution and many variables
GREP¶
0.25° · daily/monthly · 1993-01-01 – 2023-12-31
Global Ocean Ensemble Physics Reanalysis. An ensemble of several models; the member products can be downloaded from the same link
GLOBAL_MULTIYEAR_PHY_ENS_001_031
NCEP GODAS¶
monthly · 1980-01 – recent (~2 months delay)
Multi-level ocean analysis from NOAA/NCEP
GODAS (NOAA PSL)
GODAS (NOAA CPC)
NCEP products (NOAA PSL)
| Product | Description |
|---|---|
| NCEP GDAS | NCEP's twice-daily global analysis at 2.5° resolution on pressure levels, a product of its operational forecast system |
| NCEP GODAS | High-resolution multi-level ocean analysis from NOAA/NCEP |
| NCEP NARR | High-resolution combined model and assimilated dataset, 1979 – near present, 8-times daily, daily, and monthly on a Northern Hemisphere Lambert Conformal Conic grid (PSL NARR page) |
| NCEP-NCAR Reanalysis 1 | 4x daily, daily, and monthly atmospheric model output, 1948 – near present (PSL R1 page) |
| NCEP/DOE Reanalysis II | Improved version of Reanalysis 1 with fixed errors and updated physical parameterizations |
ECCO¶
Free-running ocean model solution fitted to observations by least squares (state estimate)
ECCO products
ECCO V4r4 T/S monthly, llc90 (PO.DAAC)
Northwest Pacific State Estimation (Scripps)
SODA¶
Simple Ocean Data Assimilation
SODA (UMD)
Climate Data Guide: SODA
UMD AOSC physical oceanography
OFES2¶
0.1° · 76°S–76°N · 1958-01 – 2016-12
No data assimilation. 105 levels (to 7500 m), JRA55-do forcing, M2 and K1 tides included
OFES2 description (from a published study)
We mainly adopt monthly variables output from OFES2 (Sasaki et al. 2020) to examine the long-term changes of the winter circulation in the Yellow Sea. OFES2 assimilates no observations and provides dynamically consistent variables covering the global ocean except the polar regions (76°S‒76°N) with a horizontal resolution of 0.1° and spanning from January 1958 to December 2016. The bottom topography in OFES2 is constructed with partial bottom cells using the bathymetry dataset of 1 arc-minute global relief model of Earth's surface (ETOPO1; Amante and Eakins 2009). The number of vertical levels is 105 with a maximum depth of 7500 m. The thickness of each layer is 5 m within the upper 100 m and increases to 300 m in the bottom layer. There are 20 levels within the upper 100 m and 55 levels within the upper 500 m. The 3-hourly JRA55-do atmosphere dataset is used for sea surface atmospheric forcing. Constant barotropic tidal currents of M2 and K1 generated by the FES2012 tidal model (Stammer et al. 2014) are used as the largest semidiurnal and diurnal tidal constituents. Note that tidal processes are of vital importance in shaping the three-dimensional structures of ocean circulation and ecosystem in the global shelf seas (Simpson and Sharples 2012; Timko et al. 2019; Wu et al. 2023a, b, 2025). Including tides and with higher horizontal and vertical resolutions, OFES2 could give more accurate and detailed three-dimensional responses of the water masses and circulation to climate variabilities and changes than other available ocean products in the Yellow Sea (Li et al. 2024).
Regional models¶
FORA-WNP30¶
1982–2016 · eddy-resolving
4D-Var Ocean ReAnalysis for the Western North Pacific. The first eddy-resolving reanalysis of the western North Pacific over the last three decades
JCOPE2M¶
1993-01-01 – present · 10.5–62°N, 108–180°E
Japan Coastal Ocean Predictability Experiment. Provided by ftp after an e-mail request (Fortran binary or NetCDF)
How to request JCOPE2M reanalysis data
Application Laboratory freely distributes JCOPE2M reanalysis data to any interested persons. If you accept the conditions of data use, send an e-mail describing your order: time range (maximum period: 1993-01-01 – present; 2 TB+), spatial range (maximum: 10.5–62°N, 108–180°E), and variables (sea surface height, north-south velocity, east-west velocity, potential temperature, salinity, heat flux at sea surface). They will provide the URL of an ftp site for downloading your data (Fortran binary or NetCDF) with a README document.
Contact: jcope (at) jamstec.go.jp
CORA 1.0-NWP¶
China Ocean ReAnalysis. Regional reanalysis of the Chinese coastal and adjacent seas
Arctic: CMEMS¶
Arctic tidal analysis (15-min, 3 km, since 2017-12-19, 53–90°N) and the TOPAZ reanalysis and forecast
Arctic Ocean Tidal Analysis and Forecast
TOPAZ: Arctic Ocean Physics Reanalysis
TOPAZ: Arctic Ocean Physics Analysis and Forecast
Tide & climate models¶
HRET14: baroclinic tide¶
Harmonic constants of sea surface height and surface currents for the M2, S2, N2, K1 and O1 internal tides (Zaron & Elipot)