Using machine learning to accurately simulate the oceanic barium distribution

Mete and colleagues (2023, see reference below) used Machine Learning (ML) to predict the global distribution of oceanic barium (Ba). Models were first trained to predict [Ba] from standard oceanographic observations using GEOTRACES data from the Arctic, Atlantic, Pacific, and Southern oceans. Model predictions of [Ba] were then compared with actual [Ba] data from the Indian Ocean, with the best models achieving a mean absolute percentage error of just 6.0 %. This successful comparison allowed the authors to calculate the global distribution of [Ba], Ba*, and marine barite saturation using data from the World Ocean Atlas. This approach revealed four significant findings: 1) the ocean contains 122±7 Tmol of dissolved Ba; 2) the variability in the barium–silicon relationship is consistent with the biogeochemical characteristics of both elements; 3) marine barite saturation exhibits systematic spatial and vertical variations; 4) taken as a whole, the ocean below 1000 m is at equilibrium with respect to barite. These results have broad implications, both for the modern ocean and for interpreting paleo-records of barite. A data product, which includes a global grid of predictions and the ML model itself, is freely available from BCO-DMO: https://www.bco-dmo.org/dataset/885506.

Figure: Model output showing the dissolved distribution of [Ba], Ba*, and barite saturation state (Ωbarite) in the surface of the Southern Ocean. Barium-star represents the difference between ‘in situ’ (i.e., ML model predicted) and silicate-predicted [Ba], defined as Ba* = [Ba]in situ – (0.54 × [Si] + 39.3). Barite saturation state, Ωbarite, is the ratio between the Ba and sulfate ion product and the in situ barite solubility product. The dashed and dotted lines show the locations of the southern Antarctic Circumpolar Current Front and the Subantarctic Front, respectively.

Reference:

Mete, Ö. Z., Subhas, A. V., Kim, H. H., Dunlea, A. G., Whitmore, L. M., Shiller, A. M., Gilbert, M., Leavitt, W. D., & Horner, T. J. (2023). Barium in seawater: dissolved distribution, relationship to silicon, and barite saturation state determined using machine learning. Earth System Science Data, 15, 4023–4045. Access the paper: 10.5194/essd-15-4023-2023

Latest highlights

A thorough estimate of the hydrothermal plumes on neodymium concentration and isotope oceanic cycles

Basak and coworkers investigated the influence of particulate matter on neodymium distributions in the Southern East Pacific Rise Hydrothermal Plume.

What are the drivers of the distributions of cadmium, nickel, zinc, copper and cobalt, manganese and aluminium in the Atlantic Ocean? Two papers are tackling this issue

The authors reveal that the distributions of dissolved tracers at depth in the South Atlantic are predominantly controlled by the mixing of North Atlantic Deep Water and waters of Antarctic origin…

Disentangling the sources and transport of iron in the Southern Ocean using a water mass mixing model analysis

Traill and co-workers used an extended optimum multiparameter analysis water‐mass mixing model…

A detailed investigation of iron complexation by organic ligands in the Western Tropical South Pacific Ocean

Léo Mahieu and his co-workers present the conditional concentration and binding-strength of iron-binding ligands during the GEOTRACES TONGA cruise.

Rechercher