OceanDepths: New AI-Ready Dataset Pairs Surface and Subsurface Ocean Observations
OceanDepths has launched as the first-ever global dataset that’s ready for AI use. It combines satellite data on sea surface temperature, salinity, and height with EN4 profiles for subsurface temperature and salinity. Additionally, it includes GLORYS12 ocean reanalysis data, making it useful for comparative research or multi-step learning. Despite covering over 70% of our planet, the oceans are still poorly monitored, hindering our understanding of ocean dynamics. Earlier datasets lacked standardized, high-resolution formats that linked satellite data to in situ observations. By uniting different types of data, OceanDepths seeks to enhance oceanic research and machine learning, focusing on observational precision to support AI analysis.
Key facts
- The world's oceans comprise over 70% of Earth's surface and are critically underobserved.
- OceanDepths is the first open, global, regridded AI-ready dataset of its kind.
- It pairs satellite-derived sea surface temperature (SST), sea surface salinity (SSS), and sea surface height (SSH) L4 products with co-located EN4 subsurface temperature and salinity profiles.
- The dataset is complemented by matched GLORYS12 ocean reanalysis data.
- Existing oceanographic resources are often model-reconstructed gridded products rather than observations.
- Prior datasets typically cover only a single variable or basin.
- Available datasets have resolutions too coarse for mesoscale dynamics.
- OceanDepths supports comparisons or multi-stage learning through the inclusion of reanalysis data.
Entities
—