Summary
This work presents a global monthly climatology of seawater total alkalinity derived from the GLODAPv2 quality-controlled dataset using a neural network model (NNGv2). The model successfully captures seasonal variability in AT, a key variable in the marine carbonate system and ocean acidification, with validated predictive accuracy across independent oceanographic time-series stations. The resulting spatiotemporally resolved AT product provides a resource for assessing marine carbon cycle dynamics and acidification trends.
UK applicability
The global AT climatology is relevant to UK marine science and climate monitoring, particularly for assessment of ocean acidification in UK waters and the North Atlantic. The dataset and methodology could support UK-led oceanographic research and contribute to marine environmental monitoring under UK climate and environmental commitments.
Key measures
Total alkalinity (AT) concentration (µmol kg⁻¹); root-mean-squared error (RMSE) of neural network predictions; spatiotemporal resolution of climatological fields (1° × 1° horizontal, 102 depth levels, monthly to annual temporal resolution)
Outcomes reported
The study generated a monthly global climatology of seawater total alkalinity (AT) using neural network modelling of quality-controlled oceanographic data. The model achieved root-mean-squared error of 5.3 µmol kg⁻¹ on training data and 3–6.2 µmol kg⁻¹ on independent time-series validation datasets.
Topic tags
Dig deeper with Pulse AI.
Pulse AI has read the whole catalogue. Ask about this record, its theme, or how the findings apply to UK farming and policy — every answer cites the underlying studies.