Department of Atmospheric & Oceanic Science

Ocean Climate Lab: Monthly ocean heat content

SODA3/4 Global Ocean Heat Content

James A. Carton, Gennady Chepurin, and Luyu Sun
1 Department of Atmospheric and Oceanic Science, University of Maryland, College Park, MD

Half of the sunlight that enters top of the atmosphere (341 W/m2) is absorbed at the surface of our planet. This surface energy flux is 1) stored locally, 2) advected horizontally, or 3) returned to the atmosphere. Here we present monthly ocean heat content during the 43 year period 1981-2023 based on eddy-permitting and eddy-resolving versions of the SODA ocean/sea ice reanalysis driven by meteorological forcing. Monthly anomalies of energy per unit horizontal area (J/m2) are presented relative to their 19-year average (2005-2023) over two fixed depth intervals 0-1000m and 0-2000m, and shallower than the depth of the 26C isotherm (tropical cyclone heat potential).

Heat Content Data Repository

Access SODA3/4 gridded monthly heat content and gridded monthly tropical cyclone heat potential datasets. The former are available both as annual files and as single files, concatenated and gunzipped.

Access Data Directory

Key Highlights

43-Year Continuous Record

Continuous monthly record spanning 1981 through 2023 .

Two analyses

Here you can compare SODA3 (1/4° × 1/4°) and SODA4 (1/10° × 1/10°) results

Two depths

For each analysis monthly heat content anomaly is available for both 0-1000m and 0-2000m depth ranges as well as tropical cyclone heat potential.

Sample Analysis 1

Figure 1: Heat storage rate integrated vertically (0-1000m) and averaged 2005-2024 in units of W/m2. (upper) SODA3, (lower) SODA4. Hot colors show places such as the subtropical North Atlantic that are gaining heat while cool colors show places such as the subpolar North Atlantic that are losing heat. The spatial average of these fields, shown in Fig. 2, gives estimates of the ocean's storage of the earth's energy imbalance.
SODA3/4 0-1000m Heat storage averaged 2005-2024

Sample Analysis 2

Figure 2: 5-year running time mean of the global ocean energy imbalance in W/m2 estimated using (red) 1/4-deg SODA3 and (blue) 1/10deg SODA4. Global ocean energy imbalance is presented for (solid lines) 0-1000m and (dashed lines) 0-2000m. Note that the Argo hydrographic profile observing system was first deployed for depths of 0-1000m between 2001-2005 and extended to the deeper 0-2000m layer by 2014.
five year running average of SODA3/4 heat global storage

Sample Analysis 3

Figure 3: Hurricane heat potential, defined as the heat content of water warmer than 26C, in the Gulf of Mexico during the months of (left) August and (right) September, 2005. Fig. 3 shows that hurricane heat potential is more intense (upper) in higher resolution SODA4 than it is in (lower) lower resolution SODA3. The months of August/September, 2005 are interesting because they span the time when Hurricane Katria swept across the Florida Peninsula, and made landfall at New Orleans.
Hurricane heat potential in the Gulf of Mexico during Aug/Sep, 2005

Data Specification & Access

The SODA3/4 heat content analysis fields are publicly accessible via HTTP server hosted by the Department of Atmospheric and Oceanic Science at UMD.

wget -r -l1 --no-parent --progress=bar -nd -A 'soda4.15.2_heat2000_mn_2005-2023.tar.gz' https://dsrs.atmos.umd.edu/DATA/DASH1/heat_content
Parameter Specification
Spatial Resolutions Global 1/4° × 1/4° grid (0.25° × 0.25°) or 1/10° × 1/10° grid (0.1° × 0.1°)
Temporal Resolution Monthly (1981-2023, or 2005-2023)
Primary Variable Heat content anomaly relative to 2005-2025 (HC J/m2)
Input Source A key input data set is the World Ocean Database of hydrographic profiles
Data Access Link http://dsrs.atmos.umd.edu/DATA/DASH1/heat_content

System & Mapping Methodology

SODA3/4 use sequential data assimilation to constrain a nonlinear model of the primitive equations forced by specified ERA5 surface meteorology

Validation & Quality Evaluation

SODA3/4 accuracy has been rigorously evaluated using four complementary methods:

  1. Independent Observations: Comparisons have been made to data sets such as sea level that have not been assimilated
  2. Analysis Increments: Increment statistics confirm negligible time-mean bias and no evidence of spurious trend.
  3. Intercomparison: Statistical comparisons between SODA3 and SODA4 and comparisons to reanalyses from other centers have been published.

Citation & Contact

When using DASH1.0 data in publications or presentations, please cite the primary reference:

Gennady A. Chepurin, James A. Carton, Luyu Sun, and Stephen G. Penny, SODA4: a mesoscale ocean/sea ice reanalysis 1980–2024, https://doi.org/10.5194/egusphere-2025-3810

Corresponding Author: James A. Carton (carton@umd.edu)