Mathematical Geophysics Research Spotlight

Reconstructing the Ocean's Murky Past

Scientists test whether sparse, indirect data can reveal ancient ocean chemistry and circulation patterns.

Source: Paleoceanography


For decades, scientists have used the fossilized shells of foraminifera to help them explore the ocean’s past. Proportions of different elements and isotopes in the shells give a rough snapshot of ocean conditions at the time the organisms lived on the seafloor and were buried in sediment.

However, foraminifera data are limited and difficult to obtain by deep-sea sediment coring, and the shells are not perfect proxies for ocean conditions. For example, both temperature and pH can affect their isotopic composition. New research by Gebbie et al. explores just how accurately the ocean’s past chemistry and circulation patterns can be reconstructed from the sparse, indirect evidence available.

Distribution of the stable carbon isotope ratio of dissolved inorganic carbon averaged between 2000- and 3000-meter depth.
Distribution of the stable carbon isotope ratio (δ13C) of dissolved inorganic carbon averaged between 2000- and 3000-meter depth in (top) a modern-day reference case and (bottom) a reconstruction using sparse observations of seven properties at the 492 locations of paleoceanographic data (squares). The reconstruction captures the lowering of δ13C from the Atlantic to the Pacific consistent with the known aging of waters. Click image for larger version.

To see how well they could interpret the past, the researchers tested their ability to reconstruct the present. They used a method that the lead author devised in 2014 while recreating conditions in the North Atlantic Ocean during the Last Glacial Maximum, 20,000 years ago.

First, the team tested how well this method could reconstruct modern ocean conditions, given sparse data. They limited data sources to locations around the world in which paleoceanographic data also exist, using seven variables: salinity; potential temperature; oxygen isotope ratios in water; carbon isotope ratios in dissolved inorganic carbon; and concentrations of phosphate, nitrate, and dissolved oxygen.

The authors found that this small but diverse suite of evidence accurately reproduced modern ocean chemistry and circulation patterns. The method they used outperformed another method—optimal interpolation—often used for paleoceanographic mapping.

Then the team restricted evidence to just three measurements that can be obtained from shell samples in sediment cores: cadmium concentration and carbon and oxygen isotope ratios in the calcium carbonate shells. These data reflected both the sparsity and uncertainty of the evidence available for the Last Glacial Maximum.

The sparse proxy data failed to predict modern large-scale ocean circulation patterns. These results suggest that paleoceanographers should not focus exclusively on retrieving more sediment cores from new sites. In addition, they would be well served by the ability to gather more lines of evidence—at least three additional types—from existing sediment core samples. (Paleoceanography, doi:10.1002/2015PA002917, 2016)

—Sarah Stanley, Freelance Writer

Citation: Stanley, S. (2016), Reconstructing the ocean’s murky past, Eos, 97, doi:10.1029/2016EO051923. Published on 11 May 2016.

© 2016. The authors. CC BY-NC-ND 3.0
  • T Michelle Gapinski Goodale

    ^Depth and temperature dependent multifaceted organisms are they! Pretty illustrations and images of these thermohaline **proxies! Thanks for this and recent publications from the EOS. So glad I rejoined AGU.