Global Monthly Climatology of Oceanic Dissolved Inorganic Carbon via Neural Network

A global monthly climatology of oceanic total dissolved inorganic carbon (DIC) was created using a neural network trained on GLODAPv2.2019 and LDEOv2016 data. The model uses predictor variables including position, temperature, salinity, and nutrients to produce a grid with 1°x1° spatial resolution and 102 depth levels. The dataset was published by the National Oceanic and Atmospheric Administration (NOAA) and last updated in March 2026.

Use Cases

• Modeling ocean carbon storage and fluxes based on global DIC climatology.
• Studying ocean acidification patterns based on dissolved inorganic carbon distributions.
• Validating biogeochemical ocean models based on neural-network-derived DIC fields.
• Analyzing seasonal variability in carbon chemistry based on monthly resolution data.

Strengths

• Global spatial coverage across all major ocean basins.
• High vertical resolution with 102 depth levels from 0 to 5500 meters.
• Monthly temporal resolution for the upper ocean (0-1500m).
• Derived from established source datasets GLODAPv2.2019 and LDEOv2016.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect geographic or temporal bias inherent to the training data sources.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

National Oceanic and Atmospheric Administration, Department of Commerce

Collection Method

Neural network model trained on GLODAPv2.2019 and LDEOv2016 data, using predictor variables from WOA13 and CANYON-B.

Time Range

Climatological period (specific years not stated).

Freshness

Last updated 2026-03 05 23:12:18.055874; freshness should be verified.

Geography

Global ocean coverage.